PART II
Chapter IX
Change and History
The essential feature of Aristotle's metaphysics lies in the skill with which Aristotle admits all forms of change or motion (kinesis or, better, energeia ) within the encompassing terms of invariant reality. Aristotle's was the first and most resilient reconciliation of change and changelessness the ancient world had ever achieved. The work of the Presocratics, notably that of Heraclitus and Democritus, proved too fragmentary for later use in any systematic way, and Plato's great effort (regarding the eternal Forms) has remained altogether baffling, from Aristotle's time to our own, so that it cannot really count as a fully explicit alternative. (Of course, it has inspired all sorts of alternatives.) More important is the intriguing and remarkable fact that history, although plainly a form of change, cannot be reduced to (that is, explicated, without significant remainder, in terms of) any other forms of change (those, say, that Aristotle admits). Not only that, but what we now mean by "history" had hardly dawned on the Western world until a philosophical review of the import of the French Revolution was seriously attempted. The significance of this single fact can hardly be exaggerated.
The conceptual link between theorizing about history and contingent history itself is probably unique in the chronicles of philosophy. That very fact bears in a most instructive way on the puzzles aired in the preceding chapter. For, you will remember, I claimed there that the rationale for legitimation could not be satisfactorily supplied except in historicized terms and that that signified a certain profound contingency regarding what we should understand by "reason." It now begins to
appear that the theme of history is an unusually large one, not in any sense a mere adjustment (adding a species of change) to an otherwise more or less adequate account of reality. In its modern form, the theory of history was hardly pursued in the ancient world. (You have only to read the few lines Aristotle spares history in his Poetics .) This is not to say that Thucydides's history was not a genuine history. But Thucydides's history was also conceived in full accord with the common theme of classical philosophy: that is, the archic theme that change is ultimately constrained by the changeless order of nature.
What I am claiming is admittedly open to dispute. You may insist that the theme of history was discovered by the Greeks: that the difference between modern and ancient notions of history is not due to the appearance of a radically different notion of history (a sui generis form of change) but only to changes in our conception of the world in which histories and change obtain; hence, that, since the Greeks had in fact considered (and rejected) the conceptual viability of denying invariance, they may (or must) be said to have considered as well the radical sort of history I am alluding to (the form of change that, in chapter 8, proved to be unavoidable).
I regard this last sort of resistance as a grand mistake, one that utterly obscures the distinctive features of human history. The Greeks did not theorize about history, because they did not consider the possibility that history was a sui generis form of change entirely unlike whatever forms of change might be found in physical nature.[1] They theorized instead about generic change (in nature), they subsumed history under that, and they admitted no essential distinction between mere (temporal) change and history. They supposed the chronicles of human life present us with information about change, about events that take time—in a sense of "change" suitably uniform for physical events and human affairs alike. Simply put, their view was that the temporal process (whatever its variety) was uniformly constrained, both epistemically and ontically, by the changeless order of things.
That is what I mean to deny: both (i) the necessary link between history and changelessness, and (ii) the indifference of history's real structure to the analysis of change itself, hence to the analysis of the "nature" of the things that change in the way of history. History, I say, is a sui generis form of change that qualifies only things of a certain "nature" apt for such change. The Greeks believed that temporal change (in effect, what they conveniently called history in certain accounts) applied in a uniform way to anything in (sublunar) nature —a fortiori, to
humans. Much of the modern discussion of history agrees with this part of the Greek view: that is, with item (ii) of the tally just offered. (I mean to replace it.) Plato's formula, in Timaeus, declares that "time is the moving image of eternity"; Aristotle's, in Poetics, that "poetry is more philosophical than history." Neither touches the essential issue.
The Greeks do not theorize about history, then, because they do not see the global significance of history as a conceptually distinctive kind of change affecting only distinctive kinds of entities (or, the entire world because of that). They do not see that the analysis of history bears on the distinctive metaphysics of human existence and the corresponding features of human knowledge. They theorize about human existence and knowledge all right—in terms of a generic relationship between (natural) change and changelessness—and then they admit history as a record of sequentially ordered changes from which, as in Thucydides, we may even glean political and ethical wisdom. Kinesis and energeia, in Aristotle's sense, are certainly ample notions. But the heart of Aristotle's account—which is no different in this respect from that of any other ancient theorist—concedes that the exemplars of change may be freely or indifferently drawn from the inanimate and subhuman worlds; that, I say, betrays the Greek blindness to the distinction of history.
In any event, returning to theorem (2.1)—the denial of the necessity of invariance—I now safely affirm that
(9.1) history does not presuppose a changeless order of reality.
Theorem (9.1) trivially recovers the large theme of (2.1) within the terms of reference that began to surface in chapter 8. More provocatively, I claim (but cannot yet show) that, within the scope of (2.1),
(9.2) history is a sui generis form of temporal change.
In this sense, the idea of history is perhaps the single greatest philosophical discovery of the nineteenth century. I shall come to my argument in a moment, but let me make clear its full intent. I mean to argue that
(9.3) history is the sui generis (Intentional) structure of human existence and of whatever belongs intrinsically to the world of human culture;
hence, that, at the very least,
(9.4) histories are real predicables.
Theorems (9.1)–(9.4) make a tidy set. It captures the modern notion of history—what accounts for all the fundamental differences between the modern and the ancient treatments. History, as I use the term, signifies (predicatively): (i) a unified sequence of temporal change; (ii) one that is paradigmatically manifested in human "careers" but is also found (only) in the "careers" of the artifacts of the human "world"; (iii) one that is intrinsically "meaningful" (Intentional) in a sui generis way; and (iv) one that is attributable to existing things without entailing any telos in the sequence of changes that it unites (qua history)—that is, without entailing a final cause somehow immanent and effective in the phases of an actual history. Since, as we have already seen in chapter 7, the term "career" functions equivocally as a predicate and as a referring (or individuative) expression, we should not find it unreasonable, now, to acknowledge, by the same equivocation, that
(9.5) a history is a kind of career.
Histories, we may say, are either or both predicables and individuated entities of some kind ((7.22)–(7.23).) Theorems (9.1)–(9.5) are noticeably modest, therefore, although they do indeed intrude a new theme—particularly (9.3). As they stand, except for (9.3), the set of theorems (9.2)–(9.5) hardly poses difficulties for Aristotle. It is of course (9.1) that radicalizes the rest of the set.
I have been inching forward here in a distinctly hesitant way. The obvious reason is that I take the philosophical significance of history to be of the grandest scope, to have been neglected or fundamentally misperceived by the entire Western world up to the period just preceding the French Revolution, and to have been inserted into the running account of my primer without adequate grounding. I have shown the indispensable role of historicity in resolving the puzzles of legitimation ((8.30), (8.31), (8.34)), and I regard that gain as vindicating the theorems so far introduced in this chapter. I have also shown that the doctrine of symbiosis leads, ineluctably, to the impossibility of disjoining the cultural and natural worlds—succinctly expressed in (8.35). That leads on (as I have shown) to historicizing thinking ((8.26), (8.32)).
But I have certainly not prepared the ground (in any pertinent met-
aphysical sense) for the full use of such terms as "Intentional," "cultural," "human," or "historicized": theorem (9.3) might be (wrongly) suspected of having assumed that more was safely in place than has actually been worked out. Have patience. I should not want to give the impression, having opposed Aristotle on the archic issue, that I had (also) somehow justified (or thought I had justified) the newer issues I (now) take (9.3) to implicate (but not to have established). These are clearly meant to explicate theorems (8.34)–(8.35), which, in effect, identify the cultural world as inherently historicized. That is the radical theme I have in mind. I do invoke and use the pertinent terms, but I cannot claim to have defined the world they designate. (That is the liberty I must still redeem.)
In short, I have not yet directly addressed the question of the very nature of history's structure. But if it were confined to the temporal processes of the human or cultural world (or if it were at least paradigmatically first located there), and if—as I did say in chapter 8 but did not stop to explain—thinking and cultural phenomena are in general "Intentional" ((8.37)–(8.39)), then the argument would also surely require that
(9.6) history has an Intentional structure.
This, too, cannot be more than a purely verbal formula as yet.
"Intentionality," then, is the linchpin of the entire account. You cannot fail to see, however, that Intentionality is already implicated in the earlier discussions of reference and predication, of identity and individuation, of world and universe, of text and context, of symbiosis and social construction. All that has gone before leads inexorably (but conditionally) to the promissory note that is theorem (9.6).
We are poised, therefore, for a fresh beginning. I mean to explain the distinction of history in terms of Intentionality, and I radicalize its lesson by construing history in terms of historicity —that is, the thesis that thought is itself historied, subject to horizonal constraint ((8.35)). Now, then, from an entirely different quarter—from my account of symbiosis, theorem (4.11), which, clearly, need not in principle be historicized—we cannot fail to see that combining symbiosis and historicity will yield a distinctive philosophy, one that could not have been available before the period of the French Revolution. I call it historicism, and define it simply as that doctrine that (or whatever doctrine) results from interpreting symbiosis in historicized terms.[2] Any attempt to recover invariance or
cognitive privilege within its terms is bound to be self-contradictory ((8.31)–(8.41)). Ranke's historicism is self-contradictory—or else it is saved by God's benevolence in epistemic matters (for God is hardly confined in the horizonal way). Foucault's historicism (what Foucault means by "genealogy") is not self-contradictory at all (although it is relatively inexplicit philosophically).
As I see matters, the thread of thinking in Western philosophy that best captures what I am calling historicism runs from Fichte's Jena Wissenschaftslehre to Hegel's Phenomenology, from a more post-Kantian symbiosis to Hegel's remarkable, fully historicized achievement. Foucault, in various stages of his own work, recovers what may be consistently redeemed of Kant's apriorism (the "historical a priori ") within the larger terms of the "genealogy" of truth. In short, I am leading my primer's argument in the direction of a thoroughgoing and consistent historicism.
The schema is now very trim—and rather powerful. The trick is: I apply historicity to both constructivism and symbiosis. That yields (what I call) historicism . At the risk of an oversimplification, I should say that transcendentalism (i) precludes symbiosis, (ii) accommodates constructivism (or constructionism), but (iii) opposes historicity; as a result, (iv) transcendentalism is a form of archism that treats the "subject" of (phenomenal) constructions as not itself constructed or historicized. My sense is that Kant is a transcendentalist in the first Critique, and that readers of the third Critique attempt to escape condition (iv)—Cassirer perhaps, Dilthey perhaps, certainly Habermas, possibly Putnam (at a considerable remove). They cannot succeed, I believe, if they are progressivists or teleologists (as Peirce and Habermas clearly are), if they hold that there is a uniquely correct account of the "possibility" of a would-be science or otherwise pertinent inquiry (as Kant and, seemingly, Cassirer and, in a sense, Husserl hold), if the categories of the understanding (and related structures) are not abstracted from "empirical" inquiry but (somehow) presupposed by or innate in such inquiry (as Kant's letter to Herz affirms), or if our reflexive findings regarding the "possibility" of a would-be science are, when true, necessarily true although synthetic (in Kant's sense). Externalism, as I say, precludes constructivism. Kant is hardly an externalist, therefore. But Kant interprets the epistemic pretensions of externalism within the terms of his own transcendentalism, and that is precisely what encourages externalists like Moore in their analytic "reduction" of Kant's own transcendentalism.
In what he calls genealogy, Foucault is (in my terms) committed to historicism. I should say that Hegel was as well—as a critic of Kant's transcendentalism. But if readers of Hegel insist that he is an absolute Idealist (meaning by that that even his historicism is internal to a larger teleologism addressed to the universe ), then I must part company with Hegel and charge him with a peculiar form of incoherence, since under no circumstances can a historicized or horizoned understanding confront the universe or confront it in a cognitively competent way. (That would be to reinstate something akin to transcendentalism.) But, then, I confess I find such a reading, as a serious reading of Hegel, preposterous.
I shall proceed by very small steps: it is much too easy to go astray. The fact is that the history of the concept of "intentionality" is itself seriously flawed. It was introduced in modern times by Brentano, from a reading of its original use in medieval philosophy, and it was greatly improved by Husserl. That is the form in which it largely now survives: psychologized (by Brentano) and phenomenologized (by Husserl), with almost no attention to the collective dimension of cultural life or to history or historicity—themes which I have hinted at drawing together by a historied reading of Wittgenstein's Lebensformen . In that form, it has nevertheless proved well-nigh ineliminable, despite Quine's notorious recommendation that it be eliminated from science and philosophy on the grounds that it is philosophically negligible as far as the descriptive and explanatory functions of the principal sciences are concerned. Quine was wrong, as the record shows, although I cannot stop to explain the sense of the "intentional" in Brentano's and Husserl's inquiries. (I will return to it, you may be sure. Quine himself never collected the telling evidence for his own claim.)
What, as I say, we need to notice about Brentano and Husserl is that their usage makes no explicit provision for forms of intentionality that cannot be analyzed in (or "reduced" to) whatever versions apply first to psychologically or phenomenologically apt agents (taken singly or as externally related aggregates). Brentano and Husserl treated the "intentional" (or the psychological or phenomenological) solipsistically,[3] although that was not their intention, in the sense that: (i) Brentano defined the psychological, and Husserl the phenomenological, in terms of the cognitional aptitudes of individual agents (that is, subjects ), without essential reference to any societally embedded, irreducibly collective structures, or (ii) characterized the societal dimension of thought and language as, in some inexplicit way, presupposing the prior cognizing
powers of the merely psychological or phenomenological. The argument that follows is intended to support the quite different—largely neglected—finding:
(9.7) the Intentional cannot be reduced to, or derived from, the intentional, construed solipsistically.
More provocatively put:
(9.8) the intentional, whether treated psychologically or phenomenologically, cannot be adequately analyzed without construing its own paradigmatic structure as Intentional.
The upshot of theorem (9.8) is to disallow transcendental subjects —a fortiori, transcendental constructionism ((2.13)). This explains the sense in which the transcendentalism of Kant and Husserl, although they are hardly "externalists," has been viewed congenially by analytic philosophers drawn to their particular forms of constructionism. (That is, always assuming that the transcendental can be empirically reduced.)
An important corollary for studies in animal psychology and artificial intelligence is tendered by the following theorem:
(9.9) the intentional, in nonhuman animals, machines, and the phenomena of physical nature is "anthropomorphized": metaphoric, heuristic, or otherwise encumbered by the prior analysis of the human world.
By anthropomorphized, I mean no more than that feature of our discourse about anything characterized as a "subject" (anything that is not paradigmatically human in the way of cognition and Intentionality) that is described, defined, or explained in terms that ineliminably implicate the Intentional. Theorem (9.9) follows fairly obviously from (9.7)–(9.8). So we have made an easy and interesting gain. But that gain features the controversial possibilities of (9.3) which I acknowledged a moment ago.
Let me, therefore, now define the Intentional (not the "intentional"), as a frank term of art:
(9.10) the Intentional is a collective predicable ascribed to individual human persons, selves, or subjects (in virtue of their culturally acquired competences) or to whatever artifacts or acts are conformably endowed (that is, Intentionally) by virtue of their activity; hence, the
Intentional implicates a model, or models, of rationality (in agents, deeds, or artifacts) drawn from the collective life of their enabling society.
Language, for instance, is a collective practice, whereas speech is ascribable only (aggregatively) to individual speakers. The structure of meaningful speech implicates the rationality of speakers, and the rationality of speakers is a holist projection of legitimating norms drawn from the collective life of a historical society. (I return to the question of norms in chapter 11.) Against both the Francophone structuralists (Saussure, Greimas) and the Chomskyans,[4] we may say that, paradigmatically, in natural-language contexts, there can be no disjunction between the collective (or "systematic") aspects of language and the individuated (improvisational and context-bound) aspects of speech (or the token instantiation of speech acts) or the prioritizing of one over the other. But models of rationality are themselves historicized.[5]
Linguistic "utterance" is at once the activity of individual agents and of agents endowed with collectively qualified (linguistic) powers. By the collective I mean (i) a predicable, (ii) qualifying any of a range of phenomena that includes the mental, the psychological, the subjective, the linguistic, the lingual, the active, and the cultural, (iii) that is not reducible to the solipsistic or to aggregative relations among solipsistic referents, and (iv) signifies the symbiosis between individual agents competent in terms of (ii) and whatever social conditions account (by way of natural acquisition) for their actual competence. (In effect, as we shall shortly see, the "collective" = the lebensformlich . I use it, therefore, Intentionally. I may perhaps add that I shall retire somewhat the use of the term "lebensformlich " in favor of the "Intentional." I relied on the first chiefly because it was already familiar in the context of Wittgenstein's Investigations . But its use in a historicized spirit is clearly contrary to Wittgenstein's profoundly ahistoricist intuitions. Some, therefore, may have found my freewheeling usage annoying. In any case, the "Intentional" is a term of art that has gained an advantage from association with the lebensformlich .)
Saussure's distinction between langue and parole is a disaster; Chomsky's hypothesis about hardwired "species-specific" grammars biologically prior to all forms of linguistic utterance is a completely arbitrary posit that has never been coherently explained. In particular, if "nomological necessity" proves a dubious concept, then the entire Chomskyan strategy is placed at risk for reasons more fundamental than any
empirical (or a priori ) local claims Chomsky might care to advance. Hence, if, as I recommend,
(9.11) persons or selves are, inherently, linguistically competent humans,
then
(9.12) persons or selves possess Intentional attributes that cannot be reduced to the biology, neurophysiology, or biologically emergent psychological capacities of the individual members of Homo sapiens sapiens .
From this, it follows that
(9.13) persons or selves are not numerically identical with the members of Homo sapiens, since the first possess, whereas the second do not (unless derivatively), Intentional attributes.
Furthermore, recalling the promissory intention of (7.22)–(7.23), we may now begin to reclaim the earlier debt by affirming that
(9.14) persons or selves are histories, consequently have Intentional "natures," or, lacking "natures," are only Intentional "careers";
and
(9.15) persons or selves are not natural-kind entities,
in that their "natures" do not accord (as such) with, or directly instantiate, any strict, universal, exceptionless law of nature (in particular, any law imputed to the "mere" things of the physical [or biological] world). None of this precludes, of course, our matching, as a matter of normal practice, individual persons and the individual members of Homo sapiens . They are related, but not by identity. (I address the question in the next chapter.)
The import of all of (9.7)–(9.15) may now be conveniently summarized thus:
(9.16) all and only persons or selves are subjects,
that is to say: (i) collectively enabled; (ii) individually competent; (iii) rational; and (iv) apt for language, cognition, reflexive awareness, and choice and deliberate action. Recalling the argument of chapter 4, we may now say that
(9.17) subjects are self-interpreting texts;
and that the individuated artifacts of their cultural world (artworks, words and sentences, machines, preeminently) are also texts, since they possess Intentional properties but are not (normally) subjects .[6] Thus, to attribute expressive properties to Michelangelo's Pietà is hardly to construe that sculptural "text" as a person or as possessing a psychologically or cognitionally competent "nature", but the ascription does oblige us to answer the adequational question (chapter 5)—which is normally ignored. Theorems (9.16)–(9.17) help to focus once again the peculiar failing of naturalism.
I conclude from this that Davidson (for one) is quite right to affirm something very close to this:
(9.18) there can be no strict or exceptionless (a fortiori, no deterministic) psychological or psychophysical laws,
although I also believe Davidson is wrong to suppose that, nevertheless, the mental or psychological can be subsumed under universal physical laws, simply because either the mental or psychological can be reduced in principle in physicalist terms (which Davidson actually denies) or can be treated in terms restricted to nonreductive physicalism or the supervenience theory (which he affirms).[7]
By those last terms, which are thought to yield equivalent truth-value assignments (speaking disjunctively in terms of the mental and the physical), Davidson advances the thesis that (i) the concerns of reductive and eliminative physicalisms (or similar doctrines) are to be treated agnostically; (ii) nevertheless, it is modally (necessarily) true that there can be no change in the mental or psychological without a corresponding change in the physical, and, as a result, (iii) wherever a truth-value may be empirically assigned to reports about the mental or psychological (taken "token-wise"), equivalent truth-values can always be assigned their physical counterparts. Davidson's point is that we cannot rightly speak of correlations between the "mental" and the "physical" taken type-wise, that is, in such a way as to invite a search for genuine psy-
chological laws. The reason, apparently, is that the mental is "intentional" whereas the physical is not. What Davidson means is: the mental implicates a model of rationality that is "holist"—in the sense introduced in chapter 5 (holistr ).
But, of course, if truths about the mental can be discerned token-wise, in individual cases, then there can be no a priori or principled grounds for ruling out psychophysical laws, or "type-wise" generalizations. The "supervenience" theory is, therefore, completely arbitrary (in Davidson's presentation), for, absent identity, there is no reason (and none is ever offered) for believing the modal claim being advanced. No one, in fact, has ever shown its truth or plausibility; hence, one may even suspect that it is no more than a cryptic version (perhaps computationally motivated) of the very identity theory it is said to avoid (or replace). Furthermore, on substantive grounds, supervenience is a dubious doctrine, because it both (i) advances a modally necessary connection between the mental and the physical (or analogous relations in epistemic or moral or other contexts), and (ii) is hardly as plausible, empirically, as the "many-many" principle Feigl formulated (but feared might be true).[8] The modal claim cannot be confirmed ((2.1)). The "many-many" principle holds, with respect to the mind/body problem: that, say, for any physical movement (a hand's waving), it is in principle possible that, in context, that movement may designate any of a variety of actions (signaling a turn in a automobile, pretending to signal thus, or acting the part of so signaling, among more elaborate possibilities); that, say, for any action (making a chess move), the action may, in context, obtain through or as a result of any of a variety of physical movements (a hand's pushing a pawn across a chessboard, a hand's writing a line of script [interpreted as a move], or a hand's motion flipping a switch on an electronic chessboard, among other more exotic possibilities); and that there is no algorithmic rule for determining or delimiting such linkages. (They are only Intentionally related.)
For the moment, consider the following four features of the Intentional (or cultural) drawn from what has already been said in (9.10): viz., it is (i) inherently collective rather than solipsistic, (ii) "holist" (holistr ) rather than extensional, (iii) emergent rather than supervenient, and (iv) historicized rather than naturalized. At the very least, no effort to naturalize history and culture can claim to be conceptually responsible if it fails to come to terms with all four features. Davidson addresses only one: the sense in which the mental is "holistr "—but that feature cannot stand on its own: it leads at once to the unsupported supposition
that all "holistr " idioms are a mere façon de parler . Perhaps they are. But where is the argument?
This is a very large matter. I can afford only a few remarks. For one thing, when Davidson speaks of "types" and "tokens," he means nothing more than particular causal occasions not subsumed under a causal law: mere "members" of a "set." (In chapter 10, I shall introduce an entirely different "type/token" idiom suited to cultural entities.) Second, when he speaks of "supervenience," he means a "dependency" (the term hearkens back to Moore's account of "natural" and "nonnatural" properties in the moral context, although Moore does not mention supervenience). Hence, when he says the connection between the "mental" and the "physical" is a necessary one (in the direction of favoring the "dependency" of the mental on the physical), Davidson reads supervenience in terms that are non-Intentional, non-causal, and non-analytic (logically).
At this point, I need warn only that wherever the psychological cannot be satisfactorily characterized in non-Intentional terms (frankly, already at the level of human language and thought), supervenience cannot be a convincing doctrine for, surely, the Intentional cannot be shown to be "supervenient" on the physical (in Davidson's sense). No view of the "mental" shorn of Intentional qualification can bear the weight of providing an analysis of the cultural world ((8.39)), and the Intentional is holistr (modeled in terms of rationality) rather than extensional ((9.10)). In general, the relation between the "mental" (linguistically qualified) and the "physical" is and must be mediated by suitable interpretive schemes—what Rorty has dismissed as "tertia ." The only other possibilities are the reductive and eliminative options that Davidson rejects.
In the same spirit, we may say that, if, indeed, the psychological is Intentional ((8.37)), then (on the argument being favored, which is still largely inexplicit),
(9.19) the psychological cannot be "naturalized,"
which is a theorem that catches up the gist of (8.25) and at least part of the gist of Davidson's own claim about the mental (being not nomological as such)—if (also) "supervenience" fails. (Of course, Davidson does not believe supervenience fails.) So the complexities of our topic cannot be hidden, although I must concede again that what I have so far constructed still looks a little like a house of cards.
For related reasons, I take Popper to be right in holding that
(9.20) there are no laws of history,
if he means, by law, exceptionless or modally necessary ("nomological") universals ranging over causal and cognate relations, regardless of whether we ever discern or can confirm them. But, against Popper, this is not to say that "laws" or "lawlike regularities" need be, or ever are, necessary de re ((2.6)) or exceptionlessly universal.[9] At the present time, there is a decided retreat from the once-favored dictum that the true laws of nature are nomologically [or ontically] necessary—contextless and exceptionless. Insistence along such lines now tends to be viewed as idealized (not strictly confirmable), possibly distorted for imagined explanatory gains, or relativized to assumptions that are hardly necessary themselves. Theorems (9.13)–(9.20), therefore, invite an entirely new set of questions. As long as analytic philosophies of science (notably, in Hempel and Reichenbach, more recently in Salmon) successfully persuaded general philosophy to acknowledge the high constraint of "nomological necessity," the bare admission of Intentional complexities was judged to threaten a fatal departure from the canon of rational science. The general failing of Positivism and the unity of science has changed our perception of all that.
Davidson's "anomalous monism," for instance, is a theory that could only have surfaced in the historical space between the decline of the nomological (the nomologically necessary), the extensional, the physicalistic, and the increased accommodation of all forms of intentionality. I am not saying a realist reading of the "laws" of nature is impossible, but its usual modal presumption is now surely problematic. The decline in nomological modality varies inversely with the star of Intentionality—and, with that, the rising need for an altered vision of what a science is . (For the reverse reason, the rise of interest in the supervenience doctrine is directly proportional to the decline in the fortunes of nomological necessity .)
The single most important and most strategic theorem in this process of conceptual change, bringing the attack on modal necessity ((2.1)) to bear on the archic reading of nomologicality, affirms that
(9.21) every science, being a human construction, is itself a human science and, as a consequence, is Intentional in structure.[10]
Theorem (9.21) is the result of applying (9.10) to the terms of symbiosis, once the modal reading of nomologicality proves difficult or impossible to defend. The consequence is a deep suspicion regarding the theory of scientific law and explanation, even where (as with van Fraassen and Cartwright) new hopes spring up about the prospects for refurbishing Positivism. The consequence for scientific realism, when historicity is permitted to affect law and explanation, is the clear loss of a methodological core and a clear drift toward relativism (as in Kuhn and Feyerabend). We simply do not know what "restoration" is possible now, but, in the interim, Intentionality cannot be put back in the genie's bottle. I say, therefore, that
(9.22) science and philosophy are inherently "folk-theoretic" undertakings that function only "top-down."[11]
It is reasonably clear that a theorem like (9.22) falls within the competence of any philosophical strategy that: (i) features the lebensformlich; or (ii) favors the "post-Kantian" reading of symbiosis; or (iii) construes either of these in historicized terms. What I have been pursuing, therefore, is the enormous advantage (for such strategies) of the developing philosophical uneasiness about the "modal" reading of nomologicality, perhaps the last of the great archic themes of Western thought. That is what it means to have brought all my arguments within the ambit of Intentionality. (The structure of intentionality needs still to be made explicit.)
I do not say that (9.22) is itself a (true) modal claim. I say that it is the best philosophical bet any of us can now foresee. It is of course very widely opposed in current analytic philosophy. (It is part of the motivation for invoking supervenience.) There is, however, no (known) compelling refutation of (9.22) or even a strong alternative at the present time. That in itself I take to be a distinctly telling sign of certain errant tendencies in analytic philosophy—for instance, in eliminationism (or eliminative physicalism ), which simply insists (without any argument that I have been able to find) that Quine's (and Sellars's related) rejection of the bare (ontic or realist) eligibility of any and all forms of intentionality in bona fide scientific or philosophical inquiries is unqualifiedly compelling.[12]
Eliminationism (which, of course, is itself a "folk-theoretic" posit intended to subvert the supposed relevance and adequacy of any folk-theoretic thesis within the sciences) depends for whatever appeal it has
on the canonical standing of strong nomologicality. This is as true in Churchland (who has indeed retreated from the nomological) as it is in Quine and Sellars—and signifies a deficiency of argument for the same reasons. I should add, here, that, by the nomological —taken in the strongest sense—I mean (i) that exceptionless attribute of the structure of the real world, (ii) in virtue of which changes in the world are, rightly, causally explained, which (iii) signifies that the changes that thus occur in nature occur necessarily (as in the views of Reichenbach and Salmon). The "nomological" may, consistently, be interpreted in various weakened ways. The definition given treats the "nomological" as deterministic.
By the canonical picture of science, I understand any theory, for instance the unity of science program, that holds that first-order description and explanation of what exists, also prediction and technological control, are pursued and formulated in nomological terms at least, as just explained.[13] (Science itself is inevitably more informal regarding the nomological and the complexity of what exists or functions causally, as we shall see. Effectively, any inquiry reasonably committed to the predictive, explanatory, and technological objectives of the canon, with or without the canon's modal presumptions, may claim to be a piece of science. There cannot be a strict canon, once we yield on modal necessity. Also, one cannot then fail to ponder the parallel fates of nomologicality and supervenience in the larger context of naturalism.)
In any case, if we treat the expressions folk-theoretic and top-down as epithets signifying that "worlds" are "texts" ((4.14)), because the symbiosis and holism within which we function are themselves Intentional, then it follows that:
(9.23) the entire universe is interpreted, textual, historicized, constructed—in a word, Intentional.
The closest anticipation of (9.23) may be found in Peirce's doctrine: namely, that the triadic structure of (what Peirce calls) signs "perfuse" the (symbiotized) universe. What Peirce calls triadic I call Intentional: whatever, predicatively, is (i) collective, (ii) culturally emergent, and (iii) interpretively significant ((8.32)–(8.35)). (I depart from Peirce in rejecting a "cosmic mind" and its implications for a weakened reading of symbiosis.)
I have permitted certain intriguing clues to surface as a consequence of having introduced the notions of historicity and Intentionality, but I
have not yet analyzed them directly or essentially in metaphysical terms, or justified any inference from them except conditionally. I have, therefore, put the entire argument at risk. (I may as well concede that this is the consequence of the strategy of this primer. I could perhaps have reversed the order of parts I and II, but I should then not have been able to trace the need for the distinctions of part II in the standard problems of part I.)
I have, however, given a partial analysis of history in a tally a short while ago—(i)–(iv)—explicating the sense of (9.4), of "history" used predicatively. (I ask you to have another look at that.) We have only to add to that tally a further item—(v) the qualification of (9.6) (Intentionality)—to round out the general lines of a fair account. That is certainly tidy, although it obviously adds to the baggage of an already heavily burdened conditional argument. I don't believe it adds any untoward difficulty, however; it brings together difficulties already broached, and it actually facilitates their (joint) resolution.
I can make this quite clear through the following combination of brief summary and slight advance: (i) history = Intentionally ordered change ((9.6)); (ii) the Intentional = the cultural ((8.38)); (iii) symbiotized, worlds are texted, socially constructed, intrinsically interpretable ((4.9)–(4.16), (9.23)); and (iv) constative discourse interprets our (interpretable) worlds ((4.15)). Treat this as a refinement of the earlier tally (at (9.4))—hence, now to include item (v). Given that, it is no more than a terminological adjustment to add to the former tally, the one begun at (9.4), as item (vi):
(9.24) interpretability = intens ionality.
I now draw—from, say, (8.34), (8.45)–(8.47), which concern the historicizing of truth-claims and legitimation—as item (vii) of the same tally:
(9.25) Intentionality = interpretability.
Theorem (9.25) is, in a way, the middle term for everything that is locally at stake: all the pertinent linkages involving Intentionality, intentionality, intensionality, history, historicity, interpretation, horizonality, contextedness, symbiosis, intransparency at least.
Theorems (9.24)–(9.25) are extraordinarily useful. They serve to isolate the essential puzzle confronting every serious attempt to construe
science in a thoroughly extensional way, along two lines: first, by admitting the realism of Intentionality itself (the world of human culture); second, by admitting the ontic and epistemic dependence of scientific objectivity on the conditions of reflexive human understanding (symbiosis). These are just the themes standardly neglected in analytic philosophy.
A bit of explication may be helpful. The term intensional is standardly used in philosophical contexts in two ways: (i) "intensional" = nonextensional, in that sense in which (in as generous or restricted a reading of "logical" as one wishes) logical (syntactic, formal, truth-functional) relations between terms, between sentences, between arguments, are said not to behave "extensionally" (with regard to the assignment of truth-values)—as in accord with a ("first-order") logic like that offered in Principia Mathematica (for instance, "love," not designating a transitive relationship, is said to behave "nonextensionally" or "intensionally"); and (ii) "intensions"
There are of course famous questions about the relation of "intensions" and "extensions" in sense (ii).[14] Plainly, an archic reading (Aristotle's) would claim a regular, even rulelike, connection between intensions and extensions, but any drift in the direction of the an-archic (for instance, the post-Kantian) will (as Putnam saw) call into question the supposed regularity championed by the archist. I have no need to pursue these matters here. The main thing is this:
(9.26) the Intentional is inherently intensional.
Theorem (9.26) shows at a stroke, therefore, why the ("solipsistic") accounts of intent ionality offered by Brentano and Husserl cannot but fail. The reason is this:
(9.27) the intent ional, although conceptually distinct from the intens ional, is inherently intens ional.
Brentano and Husserl both understand the intentional to designate—irreducibly—"aboutness"-phenomena. I use this locution (" 'aboutness'-phenomena") to avoid any restriction as to whether "intentionality" belongs to the mental or the subjective or the sentential, or is admissible de re or only de dicto, or is analyzed in naturalized or realist or phenomenological terms. For the moment, I merely wish to emphasize that sentences used to describe intent ional "phenomena" exhibit a characteristically complex (indissoluble) structure, which is or appears to be: (i) irreducible to nonintent ional structures; and (ii) such that those sentences behave "logically" in intens ional ways (in accord with the first sense of "intens ional" given above; for instance, "Tom has an irrational fear of horses" and "Tom believes that horses are dangerous and unpredictable" are sentences that, on a not unreasonable theory, are both intent ionally structured and behave "intens ionally"—and behave intensionally because they are intentional). (Thus, "Tom believes that horses are dangerous" may be true, whether "Horses are dangerous" is true or false and whether there are horses or not. There you have a paradigm of the "nonextensional.") The charm of this way of analyzing the phenomena before us is simply that it does not require taking sides as between Brentano and Husserl.
By aboutness, then, I mean no more than intentionality: that aspect of thought or speech in virtue of which a "mental state" is (as we say) "directed to," "intends" (in the original medieval sense), is "about" (in the sense Brentano sketches) some ("intentional") "referent" or other said to be internal to, inseparable from, significantly informing, that particular state. Whenever, in linguistic and cultural contexts, we attribute "aboutness" to phenomena that are neither "mental" nor "linguistic" in any obvious sense (for instance, the "lingual" tristesse of the Swan Queen in Tchaikovsky's ballet Swan Lake ), we automatically construe the "intentional" as "Intentional" (in the sense introduced). (I take the slippage to be philosophically unavoidable but usually not acknowledged.)
On this reading, the intent ional is intens ional for several reasons: (i) because the "aboutness" structure (in the mind or in the sentences appropriately affirming that structure in the mind, or, by a reasonable enlargement, affirming such structures in artworks or in the cultural world at large) is not reducible or detachable in any way that would
permit an "extensional" treatment of that structure (in the first sense of "extensional" given above); (ii) because the "aboutness" structure is inherently what it is, in virtue of the significant or interpretable "content" it encompasses—in effect, the meaning of the grammatical accusative—of that structure (hence, what is specified "intensionally," whether it is, or can be, specified "extensionally," in the second sense of "extensional" given above); and (iii) because "aboutness" may be assigned a realist function. Quine concurs in this: it is the reason he outlaws intentionality.
There are of course other important considerations: for instance, what sort of reality should be accorded the intentional structures of mental states (or artworks) thus characterized; or would, say, the sentential representation of the intentional be best managed by an "objectual" idiom or by a "propositional" idiom. (For instance, should we say that Tom fears "horses" [possibly, nonexistent horses] or that Tom is in a mental state such that the [internal] intentional content of that indivisible [monadic ] state is accurately represented by a proposition like "Horses are dangerous and unpredictable?")
Granting the relevance of this much, I claim, confirms (9.27) and (9.9) (if propositional modeling is the better policy—as I recommend)—and, now, confirms the methodologically more pointed theorems (9.21) and (9.22). For, surely,
(9.28) symbiosis entails the Intentional structure of our worlds and (mythically, our) universe,
which is itself entailed by (9.21). Hence, also,
(9.29) the admission of the real world cannot preclude the (semiotized) agents (selves) by which it is constituted as intelligible and interpretable;
(9.30) the real world (or worlds) cannot fail to include real Intent ional—intens ionally qualified intent ional—structures;
and
(9.31) the intentional and the intensional obtain only as abstracted within the space of the Intentional.
I take (9.31) to confirm (9.21) and (9.22) in a straightforward way. I also thereby obviate the need to raise (as with Meinong) the question of the reality or existence or "subsistence" of the "objects" (or internal accusatives ) of intentional discourse, although I certainly do not preclude the possibility that intentional discourse may be addressed to what is actual.
If, now, you recall the discussion (in chapter 3) of reference and predication, you will not fail to see that the present review of the varieties of intentionality was adumbrated there. For, on the analysis of the inherent informality of reference and "real generals," I was led to the conclusion that the structure of effective discourse is (on its constative side) "entrenched" in the structure of the consensual practices of its enabling Lebensformen . But that is what I take Intentionality to signify (on its discursive side):
(9.32) Intentionality = the consensual (collective) structure of our Lebensformen .
Theorem (9.32) bears in a profound way on the analysis of selves, but I must postpone the reckoning. I take the liberty, however, of focusing a little more sharply the meaning of the collective; that is, that feature designated by predicables that (i) are Intentional, (ii) describe societal structures as such, (iii) apply to the aptitudes (behavior and thought) acquired by persons and their artifacts in virtue of enculturation, (iv) are not reducible to the pre-cultural aptitudes of Homo sapiens, and (v) are expressible in "rulelike" ways, as in traditions, institutions, or practices.
I come, finally, to the bearing of this gathering argument on the matter of history. If history is indeed Intentional ((9.6)), and if history is a sui generis form of temporal change ((9.2)) then
(9.33) the time of history
mere physical time.
Theorem (9.33) has a special importance. It distinguishes between "physical" and "historical" time, it's true, but, more than that, it poses the entire question of the conceptual relationship between the (real) "worlds" of (physical) nature and (human) culture. That bears directly on the metaphysics of history and Intentionality, and that is precisely what was missing in the conditional liberties I had taken (earlier) in airing a number of issues involving the analysis of change. It's what I had in mind, in fact, in challenging Aristotle's short account of history.
You will remember that I argued that it was not enough to add a further species of change to the schema Aristotle offered. What was needed was a reinterpretation of how, metaphysically, the structure of history affects the analysis of change in both the natural and cultural worlds. Of course, Aristotle's archism precluded any adjustment of the sort I had in mind. But once one subscribes to (2.1) and symbiosis, it becomes quite implausible (perhaps even impossible) to treat history in ways that do not ramify through the entire universe of discourse. I regard this intuition as the fruit of the "cunning" of philosophy itself: for the puzzle of history in the modern sense was first perceived to have been posed only at the time of (or a little before) the advent of the French Revolution—just in time, so to say, to permit history to be interpreted in Kantian terms and (consequently) to permit Kantian constructivism to be historicized (in the symbiotized "post-Kantian" manner that links Hegel and Nietzsche and Gadamer and Foucault).[15]
In short, the significance of (9.33) is barely adumbrated in the distinction regarding time; theorem (9.33) is, after all, little more than a special case of the following theorem:
(9.34) cultural, but not physical, entities intrinsically possess Intentionality, but both the physical and cultural worlds are, as the cultural artifacts they are, jointly affected Intentionally.
Obviously, if (9.33) and (9.34) and similar theorems are admitted, then
(9.35) Intentionality affects the conceptual space of physical and cultural worlds in distinct ways, in ways that need not be alethically, epistemically, or ontically uniform, but in ways (nevertheless) that confirm their belonging to the same universe of discourse.
I need to make explicit the general theorem that links (9.4), which admits histories as real predicables, to the interlocking relations holding between history, Intentionality, interpretability, and the like (collected in the tally given a moment ago on the varieties of intentionality—following (9.26)–(9.27). That theorem affirms:
(9.36) Intentionality and interpretability are real attributes of existing things;
In a more than merely terminological sense—but in that sense at least—I should add:
(9.37) all and only cultural entities and phenomena intrinsically possess Intentionality.[16]
Theorem (9.37) brings before us all the puzzles that have been collecting in this chapter. For one, if cultural entities are real—in that they exist—then the theory of such entities will have to accord with the analysis of "exists" offered in chapter 6 (where Peirce's treatment of Secondness served as the decisive clue). For another, that same analysis will have to come to terms with an argument (also in chapter 6) involving theorems (6.16)–(6.21), to the effect that what exists has a "material nature": hence, in a way that precludes what (rather baldly) is usually termed Cartesian dualism . (The standard interpretation of "Cartesian dualism" holds that Descartes subscribed to it, but some interpreters claim that he did not.)[17]
By Cartesian dualism, I understand a metaphysical doctrine that claims that there are two (or at least two) disjunctive materiae in terms of which (i) particular existing things are "composed" and (ii) they may be individuated (res cogitans and res extensa, in Descartes). I take that thesis to be unsatisfactory, because (i) it permits no compositional account of the "nature" of human persons to be more than merely conjunctive (as involving the different materiae ), and (ii) it makes a conceptual mystery of causal interaction between the parts of any such composition. Psychophysical interaction (just the issue that exercised Davidson so strenuously) proves utterly unmanageable on the "dualist" reading. I may perhaps add another weakness to this tally, namely, (iii) whatever, in the disjunctive sense intended, is an individual thing "composed" (in part) of res cogitans (or, a similar "immaterial" materia: "a mind," say) appears not to accord with the strong sense of "exists" that I have already given. (Possibly, then, the dualist sense produces difficulties for individuation.)
In any event, in agreement with the arguments of chapter 7—those involving (7.33) and (7.34) particularly—a metaphysics must be supplied that eludes the Scylla and Charybdis of Cartesian dualism and reductive physicalism. The decisive strategy is as elegant as it is simple. I remind you that, in speaking of Aristotle's metaphysics, I was led to remark that Aristotle introduces hyle as an explanatory principle ("matter" or "stuff," that from which particular things are "composed"
or "formed" in the way of individuation), in spite of the fact that: (i) individuation does not and cannot depend on hyle criterially; and (ii) hyle itself is unformed and lacks determinate properties. If it were otherwise, hyle would itself be individuated, and then the need to account for the individuation of particular things would apply to hyle as well. We should be caught in a vicious regress. (Clearly, materia
What this shows is that, although the question of the ultimate composition of particular things generates the classic metaphysical concern regarding the choice between dualism and physicalism and the like, that question is really a "dummy" question. In short,
(9.38) questions of metaphysical "composition" are nothing but questions about the stock of predicables validly attributed to particular things but cast in individuative terms.
That is, nothing of a predicable nature can be reserved for "that" of which particular things are "composed" (hyle )—over and above what we do in fact predicate of them (or better, what, qua real, they instantiate). Alternatively,
(9.39) the "composition" of particulars or individuated things cannot be detailed except in terms of ulterior particulars.
There need not be, in principle, any "infimate" particulars, but there also cannot be any viable sense in which what makes a particular thing intelligible[18] —(i) its being identifiable and reidentifiable as the particular it is, and (ii) its possessing the attributes it possesses—is itself a distinct kind of "composition."
By (metaphysical) composition, I mean, therefore, that feature of particular (material) things in virtue of which they may be construed entirely as the parts of larger integral particulars ("wholes") or may themselves be analyzable exhaustively into constituent particular "parts." Hence,
(9.40) "composition" is predicatively redundant.
Aristotle's strategy, therefore, is either completely vacuous or completely misguided. Or, the "compositional" reading of Cartesian dualism is either a thesis about the conceptual relationship (in particular the reducibility relationship) between "extended" attributes and "cognitional"
attributes, or else it is similarly misguided. (One of the strangest doctrines in all of philosophy, I may add, is the theory of "bare particulars," which rightly perceived that "composition" cannot violate (9.40), but insisted nevertheless that "number" was a matter independent of descriptive predication.[19] My point, here, is to remind you of the findings of chapter 3, to the effect that reference and predication are indissoluble parts of the same constative activity. Furthermore, I do not assume that, in the real world, the part/whole relation necessarily takes a single or univocal form.)
It follows from all this that such philosophical choices as that between dualism and physicalism are really judgments about the reducibility or irreducibility of certain sets of predicables to others. The eliminationist, for instance, believes that there are no "mental" properties: whatever is real is "physical" only. The reductionist insists that the "mental" just is (a "part" of) the "physical." The dualist claims that the "mental" is as real as the "physical" but altogether different from it.
A number of extremely important matters hang on these distinctions. For one thing, to advocate materialism (not physicalism), as I have pointed out in (6.17), is not to theorize about the nature of (the "compositional") hyle that existent things (somehow) distributively share, but to make provision for their actually being existent . On the argument of chapter 6, only "material" things effectively exist, in that (as we now understand matters) only things that possess physical properties at least (the minimal run of material properties) can exert the required Secondness to count as existent. (I find this implicitly acknowledged in the use of what are called "mass nouns" ["water," "coal"] and "count nouns" ["horse," "a piece of coal"].)
But this does not mean that existent things possess physical properties only. That is the argument of eliminationism and reductive physicalism, but it is plainly a non sequitur . More than that, it is a non sequitur probably born of the dualist's own primitive notion, viz. that whatever is not (merely) physical must be altogether other than the physical. The dualist had invented a purely conjunctive form of composition, and a great many physicalists (for instance, Churchland and Parfit) have, in opposing dualism, somehow supposed that, if dualism fails, then the conjunctive composition of the "mental" and the "physical" fails, and (as a consequence) the mental must be either illusory or nothing but the physical.[20] The motivation for construing the choice between physicalism and dualism as disjunctive and pretty nearly exhaustive (with whatever accommodations may be thought necessary regarding "function-
alism" and the like) is straightforward enough: deny intentionality and Intentionality in all their forms and there will then appear to be no other choices. It's the admission of the Intentional as real and emergent in a sui generis way that makes it possible to entertain other accounts of the mental (for instance, the theory favored in this primer).
The fact is, the particular way in which particular things are made numerically one is not touched on at all by attending to what is merely predicable of them . Here, I recommend a terminological convenience: I speak of the unicity of a thing, in speaking of what makes it (conceptually) individuatable, hence denumerable, fit for numerical identity, reidentifiable, apt for reference—in terms, say, of its "nature," its "career," its "history," its "composition," or other mode of integral organization. I speak also of the unity of a thing, in speaking of that aspect of it which, in virtue of being individuated, it is thereupon effectively assigned number and nature . (Individuation and identity are quite different issues .)
Things need not be completely "unified" in order to exhibit sufficient "unity" to allow for "unicity." (This returns us to my first use of the terms "unicity" and "unity," in chapter 7, where they were introduced intuitively. I am discharging a small debt here.) Artworks, for example, may be individuated and reidentified, but it hardly follows from that that the "unity" artworks exhibit is of the same sort that physical objects exhibit. If, in fact, cultural entities exhibit Intentionality but physical objects do not ((9.34)), then (I am prepared to claim)
(9.41) cultural and physical entities cannot but satisfy different criteria of unicity and unity.
Aristotle's hyle disappoints us with regard to both unicity and unity; so does the dualist's reading of Descartes's res cogitans and res extensa . (Methodologically, the latter have nothing to do with individuation; they concern only predicables.)
We must go further. We must acknowledge the following theorem as well:
(9.42) real predicables that are not physical (or not merely physical), and also not dualistic, may still be "material" if they are complex with respect to the physical or if the physical may be abstracted from them.
By complex (in the way of predicables), I mean, drawing from (9.42), that feature of an Intentional predicable in virtue of which: (i) a physical
predicable may be abstracted from it; (ii) it itself is (in some as yet undefined sense) indissolubly "embedded" in, or inseparable from, such an abstractable predicable; and (iii) it itself is not reducible to such abstractable and embedding predicables.
I say that a predicable is abstracted from another if: (i) the latter is complex; or (ii) it is such that the predicate that designates it analytically entails the predicate that designates the abstracted predicable. For instance, "square" analytically entails "rectangular" and "being red" analytically entails "being colored," but neither of the first members of the pairs mentioned is "complex." By contrast, I say that
(9.43) the Intentional—a fortiori, the mental, the cultural, the historical, the textual—is a complex predicable.
Because the Intentional is complex, the physical may be abstracted from it—but not "analytically." I shall say, provisionally, that the physical may be metaphysically abstracted from the Intentional (i) because the Intentional is complex, (ii) because it falls among the real properties of existent things, and (iii) because it is in some sense "indissolubly embedded," as such, in the physical.[21] (I shall, in this chapter, later introduce "incarnate" as a term of art to characterize this "embeddedness" of the Intentional and to account for "metaphysical abstraction.")
What is most curious about "complex" predicables is the bare fact that there are real predicables that are complex . In this sense, physical properties are not merely "abstracted" from complex properties; they are also in their own right predicable as real properties of existing things. That is part of what it means to say that they may be "metaphysically abstracted." (Call that item (iv) of the tally just given.) Since, however, Intentional properties are (by that tally) not reducible to the physical, since the physical is real, and since there are real entities that are qualified in terms only of physical properties, I say that both physical and Intentional properties are basic (in a sense remotely akin to Strawson's—except that Strawson employs the term for individuatable entities, whereas I restrict it to predicables). Properties are basic, then, in that (i) they are possessed by existent entities, (ii) they are themselves metaphysically abstractable from complex properties, or (iii) they are properties from which other properties are metaphysically abstractable, and (iv) the entities that satisfy (i) form kinds of things that play, as such, a relatively important role in science or the explanation of nature (physical
objects, organisms, human persons). (Normally, then, "basic" properties will be specified in accord with some theory or other of the natures or sortal natures of classifiable entities. Remember, "natures" need not be constant ((7.6), (7.14), (7.29)).) I take the term "basic" to signify nothing in the way of a single compositional account of "all there is" or an endorsement of the unity program or a fixed hierarchy of some sort of the compositional "levels" of what is real.
Now, the pretty thing is that, although, speaking thus, geometric properties are (certainly) also able to be "abstracted" from both physical and Intentional properties, no distinct entities (geometric "objects") exist in virtue of which such properties (the geometric) are "basic." For the moment, I content myself with merely mentioning that the "complexity" of Intentional properties accounts for the sui generis form in which they and the entities they qualify may be said to have "emerged."[22] Clearly, the biological is also said to have "emerged" in the course of natural events from what was once lifelessly physical—without being "complex" in the precise sense assigned the Intentional. Hence, there cannot be a uniform sense of "emergence" in which whatever is real has emerged from whatever we suppose to have been the "original" state of the world. (I shall shortly pursue this matter more pointedly. In any case, I restrict the "geometric" to the predicative, as I have already done with numbers.) The arithmetic and geometric are, then, predicative distinctions only: there are no existent entities answering to them; hence, also, they are not basic or emergent in any sense. The biological is emergent, because there are existent organisms, but biological properties are not complex (although, in speaking of the informational, this may be disputed). The cultural is emergent, since there are cultural entities (persons and artworks), but cultural emergence takes a sui generis form distinct from the biological, since Intentional (cultural, culturally formed mental) predicables are complex.
By this general strategy, I avoid (for one thing) the fatal weakness of Strawson's metaphysics, namely, that, contrary to his own plan, different "basic particulars" (as he terms persons and bodies) may occupy the same place at the same time, without being "parts" of one another. I think there is nothing troublesome about admitting that what Strawson calls basic particulars may occupy the same place—and may be "parts" of one another. (Strawson's account would regard that as contradictory. But Strawson himself cannot escape that consequence.) The point of interest is that
(9.44) existent entities that possess complex properties are emergent with respect to entities that possess metaphysically abstractable properties (basic properties) as the existent entities they are.
The biological and the (merely) physical are metaphysically abstractable from the cultural or Intentional, and the Intentional is distinctly complex. All are "basic" and therefore enter into the explanation of different kinds of "emergent" phenomena. (For the moment, the term "embedded," mentioned in clarifying (9.43), is no more than a place marker.)
Theorem (9.44) helps to focus the essential problem of the reductionism that the unity of science program has favored, for it is entirely possible that: (i) biological entities (may) be construed as emergent relative to the inanimate physical world, without being Intentional (a fortiori, without being complex); and that (ii) the conditions (a fortiori, the explanation) of the emergence of the biological and of the cultural are utterly different from each other. Intuitively, the reason is this: persons (and other cultural entities) are numerically distinct from the entities in which they are indissolubly "embedded," but that seems not to be the case with biological entities. They emerge from the inanimate world, as more complexly organized, but they are not numerically distinct from other (equally numerically distinct) physical entities in the same way in which cultural entities are. (They are not thought to be "embedded" in other distinct entities.) In particular, cultural entities are inherently Intentional, whereas physical and biological entities are not; furthermore, cultural entities are complex in virtue of possessing Intentional properties. By contrast, biological entities are not thought to be complex in the same way. Biological properties are often thought to be such that physical properties are analytically entailed in them; alternatively, the biological is often thought to be an emergent mode of functioning of the physical itself. (In the most optimistic physicalist accounts, the biological will be reduced to the physical. There is no reason to think that that could ever happen to the Intentional.) By any of these strategies, the biological would not be said to be complex .[23] (There are other possibilities of course—the matter is still open. For instance, questions may be raised about "informational" properties. I shall touch on these in a moment.)
I leave the matter of what is "analytic" (as in speaking of entailment) undefined criterially: any working distinction, however informal, will suit my purpose. Quine may have been right to reject a principled dis-
tinction between the "analytic" and the "synthetic," but, in natural-language contexts, usage will always support some effective, if informal, distinction between the two. In any case, my argument requires only that, say, the mental does not, in any ordinary sense, "analytically entail" the "physical," although the "physical" may (as I say) be "metaphysically abstracted" from it. I shall, therefore, treat the mental (where it is Intentionally qualified) as complex, in accord with (9.43). That alone completely obviates dualism and also, perhaps, supplies the ground for rejecting all the forms of physicalism.
Certainly, to characterize the mental (that is, the linguistically or lingually informed mental) as complex (in the sense given) is to go utterly contrary to the canonical view. That explains why it is usually supposed that the mental must be treated either reductively, eliminatively, dualistically, functionally, or heuristically. But there is absolutely no reason to oppose thinking that the mental is (i) emergent, and may (ii) either be complex in the manner of the Intentional (the culturally emergent) or (perhaps) not complex in that sense but still inseparable from the physical in the manner of the biological—even if, there, it is anthropomorphicably modeled: emergent within the natural, non-Intentional world. I am persuaded that the "mental" is equivocal in this way: in fact, mental phenomena that are complex in the way of the Intentional are, in our world, (iii) emergent with respect to the mental taken in the biological sense. (Nor do I rule out the possibility that the mental in the first sense may also be emergent with respect to the electronic or something of the sort, as in AI theory, or it may simply not require such a dependence. It may, for instance, be, as a façon de parler, entirely "anthropomorphized" (9.9).)
At the present time, both in the biological and cognitive sciences, reference is made to informational properties—in a way that clearly signifies that they are thought to be real . Obviously, they could always be employed heuristically—as no more than "anthropomorphized" (as defined in this chapter). But it is not clear, for instance, whether, in speaking of DNA "codes," one means that the "informational" properties of the code, ascribed to living cells, is meant only as a stopgap measure against the time when those would-be properties will be successfully "reduced" to biochemical properties analyzable in purely physical terms (whatever that may be supposed to entail) or whether informational properties are "complex" properties in a sense similar to that just explained. (I strongly doubt that the latter is the right alternative.)
I do not find the matter examined in any straightforward or careful
way among the philosophers (and scientists) who regularly invoke the "informational." But this I can say: either the near-ubiquitous use of the "informational" jargon is a new-fangled version of dualism or it plainly lacks any metaphysical backing at the present time. (The same is true of the functional, which I introduced in speaking of the biological and mental when confined to biological resources;[24] that is, by way of a [heuristic] redescription in purposive or otherwise anthropomorphized terms.) Noticeably, in speaking of computers and systems of artificial intelligence, the issue is regularly finessed.
If "informational" properties were real properties, then, of course, the biological might be said to anticipate the Intentional.[25] (I doubt that they are independent properties. They are more likely to be anthropomorphized ascriptions made of the biological.) It would be an extraordinary discovery and would raise an important question about the possible reducibility of the Intentional to the biological (the genetic, for instance), even though the biological itself may not be reducible to the purely physical. I believe that this is the gist of Chomsky's thesis. The essential barrier to this entire line of reasoning is simply that it looks as if the "informational" is in all cases "anthropomorphized"—possibly an Intentional redescription of the causal—although the biological may indeed (also) be irreducible to the terms of the physically inanimate. (It is difficult to see what else it could be.)
I therefore venture the following theorem:
(9.45) informational predicates may form a subset of Intentional predicates, whether they designate (and, in designating, describe) real properties.
By informational, read as a term of art, I shall understand those predicates that include: teleological, functional, feed-back, purposive, even certain cognitional predicates applied to phenomena that do not as such invoke aptitudes at the conscious level (for instance, those of speech); or, more generously, predicates applied to phenomena that, apart from linguistic aptitudes, are characterized in representational, semiotic, signific, symbolic, or propositional terms (as among artworks). Think, for instance, of plants "seeking" nutrients. Broadly speaking, I shall regard informational properties equivocally, either (i) as heuristic designations of real properties that they do not, in so designating, literally describe, thus modeled (anthropomorphically) on the Intentional, or (ii) as a subset of Intentional properties, whether ascribed in a realist or heuristic
way (computational properties, for instance). One obvious benefit of admitting such an accommodation is that it helps to explain how it is that (certain) norms and values may be treated naturalistically—for instance, homeostatic norms. Here, naturalizing the normative is simply exposing the anthromorphized or "redescriptive" use of an Intentional idiom.[26]But that alone hardly shows that the normative can always be naturalized .
Nothing substantive follows from this terminological decision, except perhaps to draw attention to the arbitrariness of supposing that there is a clear sense in which the lawlike features of the "informational" can already be assigned (as in Dretske) or in which the functional and teleological can (with assurance) be completely "naturalized" (as in Millikan). I take these to be premature, rather unguarded pronouncements. If informational properties are causally efficacious, then, of course, the metaphysical analysis of such properties cannot be forever postponed. (I do see that the usual use of the "informational" in the cognitive sciences is meant to outflank any usage like that of my own "Intentional." But I do not find the supporting argument.) The obvious critical question asks (i) whether the informational is itself reducible to, or subsumable under, the Intentional, and (ii) whether the "Intentional" features of the informational can be assigned without implicating the lebensformlich as that from which it is itself "abstracted." I frankly cannot see how (ii) may be resolved in favor of independent informational properties. (The relation seems reversed: informational properties appear to be anthropomorphized. But the "biological" need not be.)
I venture a further thought. If the "intentional" cannot be disjoined from the "rational"[27] —meaning, say, that the concept of intentionality bears, paradigmatically, on beliefs and desires and their role in motivating action—then, if the "informational" designates the abstracted "content" of (Intentional) mental states, the informational may be construed as either anthropomorphized in terms of the intentional (as in supposing that a rabbit reacts to the perceptually internalized representation of the silhouette of a predatory bird) or as a mere façon de parler (where "information processing" is no more than an idiom for modeling the functional aptness of the rabbit's "response" to perceptual "pickup" not first characterized in intentional terms). J. J. Gibson has observed that, characteristically, rabbits respond too quickly to the "perceived" danger to have acted on the basis of first processing the supposed internal representation. If, however, we introduce the "intentional" as designating no more than the (informational) "content" of the appropriate
neurophysiological process where that process bypasses, or is disjoined from, the supposed processing of the first sort, then (and only then) will the "intentional" be able to be "naturalized." But it is not clear that the second model is not already necessarily parasitic on the first. I confess I cannot see how the intentional (or the informational "content" of the intentional) can be thus detached from the model of rationally ordered mental states, unless it is already known that reductionism or eliminativism obtains. For what is intentional or informational in the second instance? In this sense, I regard the prospects for "naturalizing" intentionality as profoundly empirical, but also as ineluctably question-begging.
The "Intentional" remains, as far as I can see, the pivotal notion. I introduced it in chapter 8, equating it with the "cultural," but, for purposes of clarity, I should offer a more explicit sense of what the principal manifestations (the "extension") of the Intentional include. Try this:
(9.46) Intentionality signifies—paradigmatically—the constative ascribability of any of a family of predicables of an intrinsically interpretive sort: viz. those regarding linguistic or "lingual" meaning, significance, signification, intensions, signs, symbols, reference, representations, expressions, rhetorical functions, semiotic import, rule-like regularities, purposes, propositional attitudes, intentions, and the like.[28]
The principal exemplars are all linguistic or "lingual." By lingual, I mean (once again) those predicables ascribed in the cultural world that, although not themselves explicitly linguistic, presuppose linguistic aptitude and, because of such aptitude, acquire, by lebensformlich extension, functions analogous to those assigned the explicitly linguistic. For example, the performance of the Marseillaise in France after the defeat of the Nazis may have "signified" the liberation of France and may have "referred" to that liberation. Michelangelo's Pietà "represents" events in the death of Jesus and "expresses" feelings bearing on the original mourning of that death. Signing a check normally "intends" the act of disbursing one's funds and, again normally, performatively "utters" that disbursing. There is no need for greater precision at this point, except to emphasize that speaking in this way is not in the least figurative. (It has a realist import—at the "level" of what is culturally emergent.) Thus, perception, cognitively construed, is also lingual, inasmuch as it
includes a propositional ingredient. (It is that ingredient that is anthropomorphically predicated of nonlinguistic animals.)
What still remains unanswered, what haunts this entire account, is the ontic question of the relationship between (physical) nature and (human) culture, between what is non-Intentional and what is Intentional. That was the question I had in mind in beginning, in this chapter, with theorems (9.2)–(9.4) regarding the metaphysics of history. The ensuing argument appeared to lead away from that original question. But now, curiously, we are actually closer to resolving both matters—and at a single stroke.
I had broached the issue of historical change and historical time as a way of exposing the peculiar limitation and distortion of Aristotle's theory of change—and, by association, the entire later history of Western philosophy and science that, down to our own day, has (largely in accord with some form of the archic vision) construed historical change and historical time as (no more than) narrative devices used in redescribing physical change and physical time in terms of our contingent interests. I should say that such conceptions mean to "naturalize" history, along the same lines I drew attention to in those well-known efforts (from Quine on) to "naturalize" epistemology (for instance, with regard to history, in Popper and Danto).[29] (By redescribing —introduced here as a term of art—I mean heuristically recasting predicates, that function naturalistically, in Intentional terms, so that they appear to be complex although they are not.)
I have given the problem of history a strenuous form by characterizing histories as real ((9.4)) and as intrinsically Intentional ((9.6)). Admitting that much, I was obliged (complicitously, of course) to insist on the distinction between the physical and the cultural and, at the same time, to begin to provide an account of the sense in which both are real ((9.41)–(9.42)). That is where we are at the moment—and that is where we were at the start of the present chapter. I think it would not be amiss, therefore, to return to the topic of historical time and change in order to claim (or reclaim) the clue by which to resolve the several questions that have remained unanswered.
Certainly, although physical time (the time of physical change) cannot be the same as historical time (the time of human history), it would be preposterous to suppose that the two were utterly disjoint. Surely we may insist that
(9.47) historical time is inseparable from physical time.[30]
The "two" cannot be identical, because one manifests Intentional features whereas, ex hypothesi, the other does not.
You can already see that the resolution of the puzzle mirrors the mind/body problem that has bedeviled dualism and physicalism, but it offers a much larger canvas: it now turns out that the mind/body problem (admitting the equivocation on the "mental" mentioned earlier) is little more than a special case of the culture/nature problem and that, as a consequence, the usual options tendered by dualists and reductive physicalists alike are too narrowly construed. That was the point, you remember, of introducing the puzzle regarding "informational" properties. (Informational properties are, or are the analogues of, solipsistic properties.)
Still, I need to cast the argument in even more inclusive terms. For the resolution of both problems (mind/body and culture/nature) concerns itself not merely with the issues of "unity" and "unicity" already broached (and with what must be the nonconjunctive [nondualistic] "composition" of the "mental" and the "cultural") but also with reconciling any would-be answer to either issue (or both) with the independent resolution of a further, very deep paradox that infects absolutely every speculation about the nature of the real world and our place in it. (We are close to the end of our labor: it will take a little more patience.)
The paradox in question is generated by the following two intuitions, neither of which are we prepared to reject: (i) there must have been a time (in the physical world) when the world of human culture did not yet exist; and yet, (ii) what we posit as that physical world must be an "artifact" of our symbiotized culture. To insist on both truths is what it means, roughly, to acknowledge the indissoluble union of realist and idealist conceptions of the world ((5.4)). The physical world, we say, is independent of the posits of our science, but its independenceis itself a scientific posit ((7.41)). (This is very close to Kuhn's conjecture, except that Kuhn does not subscribe to symbiosis.) The first collects what, in first-order discourse, we hold true (holistically) of the things of the natural world; the second legitimates, in second-order discourse, what we affirm (distributively) in the first. And, of course, first- and second-order discourse are inseparable ((1.1)–(1.4)). Consequently, there need be no self-defeating paradox there. There is indeed a puzzle, but it resolves itself rather nicely on our admitting the symbiosis of world and language ((4.8)–(4.9)), the distinction and interconnection between "external" and "internal" relations ((4.11)), and the further distinction between the constative and mythic uses of language ((8.17)).
I claim that, within the terms of an an-archic vision, there is no plausible way of resolving the (last) paradox without subscribing to the doctrines just mentioned; I also regard the problem as the mate of the paradox of legitimation I discussed in chapter 8 (as the "antinomy of history"). In fact, taken together, the two puzzles offer the strongest possible rationalization for abandoning the philosophical programs this primer opposes. I shall call them antinomies for obvious reasons: the antinomy of legitimation (chapter 8) and the antinomy of ontic priority (chapter 9). I insist, then, that
(9.48) no philosophy is valid that fails to resolve the antinomies of legitimation (or history) and ontic priority.[31]
My thought is that the two make very fair criteria for assessing minimal philosophical success at the end of this century and at the start of the next.
Now, if we allow ourselves to come this far, we begin to glimpse in a fresh way the gathering force of the entire argument applied retrospectively and prospectively. Retrospectively, it falls out at once that
(9.49) physicalisms of every sort are untenable;
and
(9.50) all "naturalizing" philosophies fail.
The reason is simply that
(9.51) the antinomies of legitimation and ontic priority are, for "naturalizing" strategies, inadmissible, or, if admitted, intractible.
This is the reason constative discourse is "folk-theoretic," as stated in (9.22). But it is also the reason theorem (9.22) imposes no restriction at all on science or inquiry of any kind—except, of course, to debar archic presumptions. If true, this would be a windfall, for it would confirm that
(9.52) the conceptual resources of an-archic philosophy are greater than, and permit the inclusion of, any of the viable resources of archic philosophies, except for presumptions of transparency, privilege, modal necessity, universality, contextlessness, totalizing, or the like.
To put the same point prospectively, that is, in a way meant to resolve middle-range problems that require a favorable answer to our antinomies, I now recommend (for the local question of historical time and change that has been before us through this entire chapter) that we adopt the following characterization:
(9.53) time is an attribute or predicate: in particular, an adverbial qualification of any of a range of predicates implicating change or persistence among things that exist.
Theorem (9.53) accords rather well with Aristotle's view of time (time as the "measure" of motion or change) as well as with contemporary views that acknowledge the difficulty of rendering its adverbial feature in accord with the resources of first-order logic. For instance, to say that Brutus stabbed Caesar is not to say what, at least in Davidson's rendering, comes out as "($x ) (Stabbing x & At-time-t x )": no, the temporal structure is clearly an indissoluble "part" of (an "adverbial" qualification of) what it is to be the continuous event or "career" that is "a stabbing." Nothing about the extensional behavior of predicates regimented in accord with a prior syntactic policy tells us anything about the internal structure of the predicables they are said to designate.
In general, I should say that
(9.54) predicates
predicables,
for one cannot say for certain how we should construe time as a predicable, from a study of the entrenched conveniences of a certain logical treatment of temporal predicates. The matter is no longer pressing, of course, if we abandon extensionality as a realistic ideal for the analysis of all natural-language discourse. (It should, however, remind us of the de re /de dicto issue.)
A much more important matter lies elsewhere. I suggest that we shall find it both economical and philosophically promising to concede that
(9.55) historical time is incarnate in physical time.
What I mean by "incarnate" is this. A predicable is incarnate in another predicable if: (i) it is real; (ii) qua real, it is complex—that is, Intentional and indissolubly bound to some real physical predicable ("metaphysically abstractable" from it); (iii) it is "emergent" with re-
spect to that ("incarnating") predicable—that is, not reducible to, or explicable solely in terms of, the causal role of its incarnating predicables; and (iv) it is "basic." Hence,
(9.56) cultural (or Intentional) attributes are incarnate attributes.[ 32]
This is a very strong proposal. It affords, I think, the only reasonable general way of construing historical time and change as real, as Intentional, as coherent, and as in accord with an an-archic view of the world . It steers a middle course between reductive physicalisms and irresponsible dualisms—with respect both to the mind/body problem and the culture/nature problem. It is in this sense a fair specimen of a very large family of cognate solutions, including those bearing on the ontic relationship between persons and organisms belonging to Homo sapiens sapiens, actions and physical movements, machines and (assembled) materials, speech and (uttered) sounds, paintings and (applied) pigments, cathedrals and ordered stones, mental states and neurophysiological states, history and physical change, and the like. I claim that, granting whatever logical differences may arise among these and similar pairings, they all require the admission of the incarnating relationship: that particulars of the relevant kinds (all and only "culturally emergent") exhibit their characteristic "unity" and "unicity" through and only through their possessing incarnated (Intentional) "natures."[33] The solution, therefore, dares to invite comparison with Aristotle's account of ousiai and, as I say, is more directly responsive to questions of individuation and numerical identity than is Aristotle's solution.
I cannot pursue these developing lines of inquiry here. What I want to make clear, however, is that incarnation is only one among an array of possible modes of organizing the intrinsic, integral, and entire structure of entities of different kinds, with regard to their being individuatable and reidentifiable as the particular entities they are. (I shall, in the next chapter, introduce the term "embodied" as a term of art answering, in individuative terms, to what "incarnate" signifies predicatively.)
It is, I think, worth noting that, in current analytic philosophy, although the problems of numerical identity and reidentifiability have received considerable attention, much less attention has been given to what it is to be a particular thing and how particular things may be individuated if their natures are not (sortally) constant (either de re or de dicto ). The solution I have barely sketched is intended to meet both of those questions. The general lines of the argument have been laid out
chiefly in chapter 7, but now we have a way of construing the integrity of particular things of certain kinds (cultural entities) in a way that is congruent with what has already been said about reference and predication and individuation and identity.
For convenience and clarity, let me add a few terminological distinctions. By integrity I mean that feature of particular things in virtue of which their "unity" and "unicity" are (i) accounted for ontically and (ii) preserved epistemically. I have also said, in clarifying the notion of "incarnation," that incarnate (Intentional) predicables are emergent with respect to their incarnating predicables. I cannot at the moment do full justice to what I mean by "emergent," because, on the usage I intend, emergence (like causality) cannot be applied to predicables alone, that is, without attention to the (existent) entities that exhibit them, and I have not yet introduced (I shall in the next chapter) an account of "cultural entities" that, as "emergent," possess (suitable) emergent properties. But what I can say is this. In speaking of cultural emergence, I mean to speak only of things of certain kinds (and their properties), that appear in the same "world" in which physical and biological entities appear: that is, entities uniformly subject to all the resources of constative discourse.
Of these, I say that cultural or Intentional entities (possessing incarnate properties) are emergent, in that: (i) their existence and "generation" cannot be accounted for, causally or in any other way, in terms of the existence and the causal (or other generative) powers of the (non-Intentional) entities with respect to which they are emergent; and (ii) their existence and generation can be accounted for, causally (or in other ways), in terms of other entities and their causal (or other generative) powers, if and only if they belong to the same emergent "level of reality." Intentional entities are culturally emergent, then, in that the mode of their emergence and generation is specific to that level of reality at which their "integrity" is and only is preserved, viz. the cultural. (I take the claim, of course, to be a "bet," in the sense already explained.) Furthermore, its admission captures what is essentially meant in speaking of the hermeneutic circle . "Level of reality," of course, borrows, parasitically, the standard usage of the unity of science movement, but rejects its presumption. All Intentional predicates belong to the same level of reality in being Intentional, for descriptive and explanatory purposes, and they belong to a higher level of reality than physical predicables, in the sense that physical predicables may be "metaphysically abstracted" from them. (These are meant as "indicative" distinctions fitted to a promising science.)
These categories are not terribly instructive as they stand. The reason, of course, is that we need to know just what the nature of cultural entities is in order to know specifically what mode of emergence is pertinent to their integrity and what peculiarities such entities actually exhibit, with respect to reference, predication, individuation, identity, and historical change. For example, I hold that a Dürer print (Melancholia I, say) exists in multiple token prints, each with its own local history (or career) and that it may acquire new Intentional properties as a result of its interpretive history, which may (over historical time) actually "generate" properties ("in it") that could not have emerged at some earlier time.
There is extraordinarily little discussion of such complications in analytic philosophy, and what there is in continental European philosophy (in Gadamer, for instance) is, however, intuitively instructive, almost indifferent to the resolution of the kind of question I am raising. I am persuaded that the enormous complexities of the cultural world have been pretty well ignored in analytic philosophy, simply because the archic temperament (as well as a strong externalism) has held sway for centuries and because, within its terms, the analysis of physical objects has long been deemed exemplary for everything that exists and is real. (All that is doubtful now.)
Even without a full account of cultural emergence, certain very strong findings may be drawn from what has already been said. The key notion regarding emergence is that of the limited explanatory power (of whatever kind of explanation we admit) of theories whose explanantia are restricted to whatever level proves sufficient for the description and explanation of (the non-Intentional) phenomena that form the (incarnating) ground relative to which emergent phenomena are acknowledged to be emergent. If so, then the sanguine hopes of the Positivists and those who have supported the various forms of the unity of science program are doomed, for
(9.57) the sciences of the cultural world cannot, in principle, be modeled epistemically or methodologically on any science whose own epistemic model admits truth-claims addressed only to a non-Intentional world; hence, they cannot be modeled on the physical sciences—as, in the unity of science program, they are.
There cannot be a unified science (in terms that permit every "higher-level" phenomenon to be explained in terms of the phenomena of
the "next-lower" level and the bridge laws linking the two levels) if "culturally emergent" phenomena are genuine. (Bridge laws are rules of translation linking the phenomena of two distinct "levels" of reality by way of the nomological regularities at each level [or at least at the "lower" level]. The very admission of such laws requires adherence to something close to the unity program.) There are (I admit) more "optimistic" conceptions of emergence than mine (Bunge's view, for instance) that attempt to restore methodological unity in the unity theorist's way, but I know of none that admit anything like the strong sense in which I have characterized cultural phenomena as Intentional.[34] I am prepared, therefore, to construe (9.57) conditionally. (I return to the question in chapters 10 and 11.)
Still, if, on the account so far given, the world is indeed "texted" ((4.14)), if all conceptual schemes are indeed interpretive ((5.2)), if everything that may be said to exist or be real is indeed a social construction ((7.33), (9.21)), then it cannot fail to follow that
(9.58) all the sciences are sciences of the human world,
in two senses: (i) in that what they examine include phenomena that inherently possess Intentional attributes or natures; and (ii) in that whatever attributes or natures are ascribed the things the sciences examine (even things that lack Intentional natures) are ascribed only on the strength of what is ascribed in accord with (i). If one accepts (9.58) in this way, then (I say) one is committed to a folk-theoretic conception of science and knowledge in general—and in particular, committed to the view that all that is real is "socially constructed" ((7.33)).[35] (I return to the "folk-theoretic" in chapter 10.)
The irony remains that, if the opponents of the an-archic vision are right in thinking that the model of methodological rigor they favor cannot be effectively fitted to cultural phenomena construed in Intentional terms, but wrong in thinking that the Intentional can be reduced to the non-Intentional (or eliminated altogether), then, on the argument, there is no viable model of science to be had or else the opponents of the anarchic vision have seriously misrepresented the successes of the sciences and the rigor they actually require. It may even be true, for instance, that those successes have been gained by means of a careful simplification and idealization of the complexities of the Intentional; if so, that very fact would confirm (rather than disconfirm) the need for a new picture of methodological rigor. In short,
(9.59) the physical or natural sciences are inquiries abstracted from, and within, the scope of the human sciences;
and
(9.60) whatever fundamental explanatory role the physical sciences may rightly claim is rightly claimed within and only within the competence of the human sciences.
The sense of "abstraction" in theorem (9.59) is clearly nothing but the conceptual counterpart of "metaphysical abstraction" ((9.44)). The nerve of the entire argument initiated in chapter 8 and running through the present chapter, focused for instance at theorem (9.6), is plainly this, that
(9.61) real Intentionality cannot be naturalized.
I have made something of a campaign out of exposing the inadequacy of naturalism. But I urge you not to lose sight of the fact that I have also been trying to steer a middle course between naturalism, on the one hand, and transcendentalism, on the other. My thought is that human subjects, apt in the construction of the phenomenal world, are themselves part of that construction, as a consequence of symbiosis and historicity. (That was the point of the antinomies mentioned in chapters 8 and 9.) If, now, you consider the import of such strong theorems as (9.23), (9.32), and (9.37), which link these themes to the Intentionality of the cultural world and to the existence of selves or subjects, you cannot fail to see that we are entitled to draw the analogue of (9.61) for transcendentalism as well: viz.,
(9.62) real Intentionality precludes transcendentalism.
Chapter X
Mind and Culture
It is one of the failings of contemporary analytic philosophy to have misperceived, or neglected to explore with care, the conceptual relationship between the "mental" and the "cultural." This may in part explain the perennial prominence of Descartes, Hume, and Kant, for there is hardly a suggestion among these three, extraordinary philosophers though they are, of the irreducibly social, cultural, historical, collective, artifactually constituted and variable "nature" of the mental—among humans. They are, of course, the three most influential early modern theorists studied again and again by contemporary analytic philosophers. They are the progenitors of twentieth-century "externalism."
Beyond admitting the dialectical play of the profession, it would not be unfair to claim that Anglo-American philosophy has remained remarkably loyal to the thesis that the best account of the distinctly "human" and "mental" is bound to be cast in terms of whatever proves, individually and aggregatively, to be the shared, "endogenous," species-specific endowment of Homo sapiens (in effect, the solipsistic resources of our biological nature). By the solipsistic, I mean that feature of any theory of cognizing competence that accounts for all relevant aptitudes in terms of an initial biological, genetic, precultural, neurophysiological, or subjective endowment and whatever improved skills its exercise among other similarly autonomous and similarly endowed creatures may yield. (This extends the sense of the term assigned in chapter 9.)
I take this to be a fundamental mistake . It obscures, if it does not deny altogether, the apparent truth that
(10.1) the attributes of cultural life, particularly those that bear on our discursive and cognitive powers, cannot be analyzed or adequately explained in terms only of any merely biological, or biologically restricted, mental or psychological endowment; or, indeed, in terms of the cognitional powers of apt subjects distinct (as in Kantian or Husserlian ways) from the physical or biological, but known to have certain intrinsic rational or epistemic powers (solipsistically).
Theorem (10.1) is a philosophical bet directed against both naturalists and transcendentalists. No doubt the gifted members of Homo sapiens do "become" persons, but it would be an obvious non sequitur to infer from that truism that it must also be true that the Intentional or cultural must be analyzable or explicable in terms of the biological. There's a deep mystery there, and, also, an equivocation. The bet cuts both ways, challenging reductionists and anti-reductionists alike. "Persons" emerge initially, originally, by internalizing the linguistic and lingual aptitudes of their encompassing society (Lebensformen ). Homo sapiens has the capacity for that, but I shall argue that the change is an (ontically) emergent change—one, however, that entails no biological loss and no subjective gain that is not grounded in the biological.[1] (There's the mystery.) If you grant the point, then, of course, physicalism becomes impossible to defend and we have no way of guessing at the historical possibilities of our own conceptual powers.
The telltale clue to the puzzle lies in the plain fact that the restriction of the initial resources of the human psyche is—in Descartes, Hume, and Kant—never obliged to ignore the actual features of the cultural world.[2] No, the latter is taken to be as rich and complex as you please; it is only the former that are made as spare as possible. This accounts, for instance, for Locke's marvelously candid puzzlement at finding it impossible to discern in the sensory impressions that strike his tabula rasa a sufficient clue as to the source of the idea of substance (a certain je ne sais quoi ) that he realized he could not do without.
The (solipsistic) impoverishment is never perceived, it seems: for nothing that should be acknowledged in the cultural milieu is actually denied admittance to our speculations; it is only the cognitive presumption governing the explanation of all this that remains intransigently spare, viz. that whatever we discern in ourselves and the world may be accounted for, epistemically, in terms restricted to the "mind's" original biological endowment feeding (as cleverly as you please) on the incoming data of the surrounding world. In effect, a solipsistic analysis of
linguistic and lingual intelligence treats the culturally complex or Intentional features of human life as supervenient (tacitly, although in the sense supplied in chapter 9) on whatever of the "mental" is first defined biologically—or in some analogous ("nativist") way. (In terms already supplied in (9.14), it wrongly supposes that the "subjective" can be analyzed entirely in terms of a biological (or nativist) model of the "mental.") By endogenous, I mean any putatively original source of the mental, the cognitive, the subjective, the rational that is either species-specific (restricted to the membership of Homo sapiens but characteristic of its aggregated members) or otherwise native (or nativist ) to human subjects (taken singly or in some sense prior to social or cultural learning or exchange). The endogenous, accordingly, ranges over naturalistic and non-naturalistic claims: for instance, Chomsky's, as well as Kant's and Husserl's.
In Locke and in Hume, for instance, language is essentially a convenience for recording the prior work of pristine thought; in Kant, the intelligible structures of the world are imputed to it as the gift of a nativism ample enough for what we claim to discern; in Descartes, God, very much offstage as a deus ex machina, endows native human reason with whatever one requires for understanding the various parts of the Creation. In analytic philosophy, there is almost no concern to secure a sense of the "adequation" between the native powers of Homo sapiens and what their exercise is said to reveal about the environing world or the cognizing self itself. You will find the exemplary evidence in Fodor's "nativist" analysis of concepts, which is a kind of Cartesian—even Platonist—recovery of Locke's failed project.[3]
My point is this: on pain of paradox or incoherence,
(10.2) the cognizing powers of the "mind" and the cognized features of the "world" must be adequated to one another.
Analytic philosophy largely ignores this issue—which, in their separate ways, Descartes, Hume, and Kant ingeniously address. By adequation (both ontic and epistemic ), I mean the specific "matching" (equilibration ) between the conceptual powers we impute to ourselves (apt for discerning whatever may be discerned in the world) and whatever (we claim) we actually discern.
The formula is not as vacuous as it may appear, although it is quite true that different criteria of "adequation" are bound to be internal to the theories this or that particular philosophy favors. (It is precisely this
menagerie of ideas that needs to be collected and displayed.) Thus, already in Plato, the theory of Forms (and the coordinate doctrine of the soul's "recollection") "adequates" our actual discourse about justice and the good life and the conceptual resources said to make that possible. Predictably, the more powerful the role of the Forms, the more impoverished the reckoning. Similarly, in Kant, once the phenomenal world is seen to be spatially and temporally organized, the cognizing mind (Kant thinks) must be natively endowed with the capacity to impose spatial and temporal structures onto inchoate experience and then to discern those same structures in the experienced world thus constituted. Neither in Plato nor in Kant, however, do the resources of collective cultural life play an executive role in our cognitive competence.
The telltale reason seems to be this: the cognitive powers of the human self or "subject" are assigned, in effect, to account for (or match) whatever we take to be the content of our beliefs and knowledge. Even where some genesis or biography is offered that apparently acknowledges the priority of epistemic considerations, as in Descartes's Discourse and Locke's Essay, the labor seems no more than an artful way of abstractly matching once again an account of what native mental powers would be enough to account for science and intelligence. There is hardly a serious attempt, until we reach the post-Kantian Idealists, to think of our conceptual and cognitive resources in terms of the particular culture in which our developing aptitudes first take form. The essential point may be put this way: there's no question that, moving from Descartes to Kant, cognizing selves or subjects are admitted to exist, but their existence as the cognizing agents they are is never really treated existentially, that is, in terms of the cultural forces they respond to in becoming (emerging as ) the apt agents they become. One finds the latter theme in Hegel, Marx, Nietzsche, Dilthey, Heidegger, Adorno, Dewey, Gadamer, and Foucault. You cannot find it in Russell, Moore, Quine, or Davidson, or in Brentano, Husserl, Frege, Carnap, Apel, or Habermas. They might as well be pre-Kantian thinkers. In a way, they can be considered as such—if you disregard their late nineteenth- and twentieth-century projects. (By existential, incidentally, I mean no more than that aspect of the historical formation of our careers as cognitive and active selves in virtue of which our interests, convictions, and categories of understanding are what they are.)
Viewed abstractly, adequation must be trivial. It is only in the details of the match required that we mark the master philosophies of the tradition. We build our picture of the "mind's" or cognizing "self's" con-
ceptual powers by "adequating" our theories of mind or self to what we say we know. Nothing could be simpler or more difficult. Thus, cleverly but disappointingly, Aristotle treats the highest powers of reason (nous ) in a completely ad hoc and undeveloped way: whatever the changeless structure of reality is said to be, the intuitions posteriorly assigned as the exercise of competent nous cannot fail to be (trivially) adequate to its task. Nous has no interesting (existential) structure of its own. It "mirrors" (or, better, "receives") the fixed "forms" of things: its "actualized" power is no more than its "passive" capacity to be undistortedly "informed" by those same forms. Nous is nothing but a deus ex machina .
Aristotle's theory is certainly "adequated" in the formal sense, but completely vacuous as far as its scientific function is concerned. Once we call the archic thesis into doubt, we cannot suppress the fact that Aristotle has even less to say about the themes of (10.1) than the philosophers I mentioned earlier. (I should add that, by equilibration —the term is borrowed locally from Goodman and Rawls and is now comparatively standard—I mean no more than a reciprocal adequation, a conceptual reckoning regarding what theorem (10.2) requires at both poles of the cognitive process, taken together.[4] (Equilibration makes very good sense if we accept the terms of a lebensformlich symbiosis; it makes no sense at all if the cognition of the world is construed in terms of an "externalist" competence alone. That is precisely what is so suspicious about Aristotle's and Plato's accounts.)
As I see matters, the most strategic (certainly not the only) puzzle adequation must address concerns whether the Intentional (especially its collective and interpretable features) can be "adequated" (as, in effect, Descartes, Hume, and Kant suppose) to the initial biological, or solipsistic, or species-specific, or endogenous, or transcendental powers each assigns to cogniscent selves. I should say straight out that I believe that cannot be convincingly done. But the point is, it is the litmus for every viable theory of the mental and cultural. It marks the difference between the mind/body problem and the culture/nature problem.
One strategic question that cannot be addressed in terms of Aristotle's theory, but is nonetheless important in contemporary inquiries, asks whether human conceptual and cognitive powers can, with the acquisition of a culture, be greater than (that is, not "reducible" to) any of its supposed powers at birth, or whether any supposed improvement or enlargement of those powers is anything but a skilled use of our first fixed biological (or "subjective") powers combined one way or another,
so that the mind may be taken to be a "closed system." Innatism or nativism (sometimes also called rationalism ), professed (in different forms) by Chomsky and Fodor and Descartes, holds that there is no increase in human conceptual powers beyond some original (modularly isolated) endowment (genetic or Providentially assigned), and that, therefore, any apparent such increase can and must be accounted for by (what in effect are) the innate rules of operation of the mind or reason or sensorium or computational powers of Homo sapiens . That is certainly what Descartes had in mind, speaking of the "natural light of reason."
Piaget, of course, opposed Chomsky's innatism, arguing that the conceptual powers of developing children actually changed, actually developed, increased, became "greater" (emerged at a "higher level"), in virtue of an organism's continual "interaction" with its environment (including its cultural environment). In effect, Piaget's argument (also Waddington's) held that our conceptual powers could not be accounted for reductively. Nevertheless, in his own "structuralist" theory, Piaget supposed that there was a necessary and invariant sequence manifest in the more and more powerful stages of the development of the conceptual capacities he mapped—which (as Chomsky rightly observed) do not really escape (as Piaget believed they did) the larger constraints of innatism. (For Piaget, unaccountably, the environment merely triggers an innate developmental capacity. This was just the point about language acquisition that Vygotsky pressed against Piaget. It anticipates my own theme.)
My own view is that some concession to innatism (but not transcendentalism) is entirely consistent with admitting the social construction of persons or selves. All that is needed is a division of labor, so to say: the admission (for instance, with Bruner) of some initial "hard-wired" endowment open thereafter to alteration or enlargement or supersession in ways that cannot be convincingly assigned or accounted for innately. The advantage of this option (which is not incoherent, as Piaget's may well be) is that it accommodates two essential intuitions: (i) that the mental does indeed belong to the biological endowment of humans; and (ii) that man's conceptual and cognizing competence is culturally generated and does vary in power and structure, at the level of cultural emergence, from society to society and from one phase of human history to another. We may concede that humans have innate cognitive powers, but we cannot rightly say what they are, what limits they impose, what "rules" they function by. The truth is, whatever we claim about our
innate (Intentional) endowment is itself an "anthropomorphized" conjecture viewed from the reflexive level at which we discover our actual aptitudes ((9.9)–(9.12)). Hence,
(10.3) innatism and social constructionism are not in principle incompatible, but they cannot apply coextensively;
and, of course,
(10.4) the "mental" is, in any pertinent generative sense, jointly biological and cultural.
There is, I think, a very simple way to accommodate (10.3) and (10.4), namely, by admitting that
(10.5) our culturally acquired (culturally emergent ) conceptual and cognizing powers are incarnate in the biological structures to which our innate mental capacities are directly ascribed.
In that sense,
(10.6) our culturally acquired aptitudes are incarnate in our innate endowment.
What we need to remember is that
(10.7) our innate mental endowment is and can only be inferred from the vantage of our reflexive (linguistic) competence (anthropomorphically).
Since all forms of physicalism fail ((9.49)), and since we are committed to symbiosis ((4.9), (4.11), (4.12)), there is no independent way to fix our innate endowment (whatever that may be), except by inference from what we ascribe to ourselves as our linguistic, lingual, and prelinguistic competence. That is,
(10.8) there is no direct perceptual access to our innate mental endowment.
We cannot query infants, and we cannot say "what it is like to be a bat" except from the human vantage: bats have no idea.[5]
I may as well add that we cannot ask the historical past—those who lived in the past—what they "thought" or "meant" or "intended" in the past, except, once again, from our present vantage. That is the insuperable lesson of Gadamer's hermeneutics and (if I may press the point, for provocation) the unacknowledged lesson of the better part of Quine's notion of "analytical hypotheses."[6] For what Quine concedes, however thinly, is the ineluctability of contingent tertia . What all this means is this:
(10.9) our innate mental endowment cannot fail to be "anthropomorphized."[7]
We must remind ourselves that our intuitions about the "mental"—items (i)–(ii), presented just before introducing theorem (10.3)—are no more than special cases of the deeper "antinomies" of legitimation and ontic priority explored in chapter 9. It is for this reason that cognitive psychology cannot be naturalized ((9.19)) or explained in transcendental terms ((9.62)).
We are moving rather quickly here. But, before I venture too much further, I should like to recover a connective theme (from what has already been said) that may easily be overlooked. The notion of "adequation" mentioned a moment ago has two foci, the significance of which requires that they be viewed together: one features the theme of the ontic resemblance between the conceptual powers of cognizing subjects and the cognizable features of the objective world; the other, the emergent lebensformlich (constructed, collective) nature of human cogniscence. The first is the theme common to Berkeley and the post-Kantian German Idealists; the second, the effect of a late Hegelianized reading of themes in accord with Wittgenstein's notion of "forms of life" (as via Gadamer and Foucault). What needs to be emphasized is both that these two themes are nearly completely absent from the "externalism" of current analytic philosophy and that they are both serviced by the doctrine of symbiosis. I do not believe any philosophy can be convincingly fitted to the puzzles of our age that does not embrace symbiosis.
There are several distinctions involving the analysis of mind that contemporary philosophy is plainly wrong about or scants. Theorem (10.1) marks the most important and the most neglected of these. It confirms that, if one treats the mental or psychological in terms restricted to some initial biological, genetic, or "nativist" sources confined to Homo sapiens sapiens, we cannot but fail to provide for the full range of the
"mental" in humans . For what is essential to being competent in the human way unavoidably involves linguistic and lingual competences. But these are specifically lebensformlich, collectively defined and, hence, impossible to analyze in terms of merely solipsistic or biological resources. There's the reductio of innatism. For,
(10.10) the lebensformlich is ontically distinct from the physical and biological.
(Wittgenstein would never have allowed this way of speaking, of course, or my other liberties. But it doesn't matter. In the same spirit, I can see how easy it would be to convert Plato's Forms into a metaphor for what I call the lebensformlich: I don't find the suggestion in Plato either.)[8] I should add that this explains the ingenuity of Chomsky's general strategy: Chomsky converts the Intentional—the culturally emergent—into the biological or nativist. He claims that there is "no other possibility" to consider, but he nowhere explains how the conversion is possible.
I have already claimed—through the running argument of chapters 3 and 8 and, explicitly, in chapter 9 where I construe the Intentional as inherently collective, in (9.10), and relate it to the consensual life of aggregated humans, in (9.32)—that the cultural, the historical, the interpretable, the linguistic, the discursive, the intelligible cannot but be (or incorporate) the formative powers by which we first "emerge" as persons or selves ((9.11)–(9.15)). I inferred (in the same context) that "solipsism" was inadequate, in (9.7); that the intentional (in humans) was Intentional, in (9.8); that the intentional among nonhumans was "anthropomorphized," in (9.9); and that, as a consequence, persons were not numerically identical with the members of Homo sapiens, in (9.13). These considerations justify my having introduced the "incarnate." For, otherwise, the admission of a numerically distinct organism not (numerically) identical with the person or self "housed" (somehow) in the same (living) body that is that organism would breed intolerable paradox.
I meant to introduce in this connection a range of distinct options regarding "unity" and "unicity" by which to ensure the "integrity" of the entities in question, including (i) bare numerical identity (as in Williams's identification of persons with human bodies), (ii) composition by mere conjunction (as in Descartes's dualistic union), and (iii) whatever, regarding entities, would be the substantive analogue of "incarnation" regarding predicables. For the moment, I merely mention that
the counterpart of incarnation, involving entities, I call embodiment. I do not say the tally just offered collects all possible modes of "integrity," but it does indicate the truth of (10.1) and the reasonableness of (10.2). And, that signifies that analytic philosophy has done poorly in its account of minds and selves.
Let me make clear the benign paradox of (10.1):
(10.11) persons or selves are numerically distinct entities possessing "natures" that exhibit collective structures.
By a term of art, then,
(10.12) persons or selves are embodied in the members of Homo sapiens, just as their cultural "natures" (their encultured properties) are biologically incarnate .
I do not hold that the individual members of Homo sapiens utterly lack mental capacities (although, as between Chomsky and Piaget, and between Chomsky and Bruner, there are various ways of reading the pre-linguistic competence of infants). My point is rather that
(10.13) the Intentional "nature" of humans marks what is not innate about them—and what, by definition, is essential to their being persons or selves.
As a consequence, in having failed to provide for the adequation between our culturally "collective" powers and our biologically (or otherwise endogenously) solipsistic gifts, a considerable part of contemporary philosophy has utterly failed its calling. I implied (just above) that that failing obtains in several distinct ways: (i) in supposing that our collective powers may be analyzed and explained solipsistically; (ii) in supposing that our lebensformlich powers may be analyzed and explained in terms of the innate competences of Homo sapiens (or pre-cultural subjects); and (iii) in supposing that the historically horizoned powers of human selves may be analyzed and explained in terms of the ahistorical powers of some pre-cultural (or transcendental) state. Failure along these lines signifies a failure to have adequated our conceptual and cognizing powers to our "nature" and to what we all affirm as the intelligible features of the world. Analytic philosophy, I say, has favored
these impoverishing commitments within its various externalist programs.
If, now, you look back to the account of reference and predication offered in chapter 3—look back to theorems (3.8)–(3.12), which affirm that both acts are entrenched in the Lebensformen in which they function—you will see at once the argument for supposing that, if, indeed, world and language and subject and object are symbiotized ((4.8)–(4.11)), then,
(10.14) cultural entities intrinsically possess collective attributes.[ 9]
For instance, a painting by El Greco, said to be baroque in style, is an individual thing whose salient "style" belongs to it only in virtue of its being a variant of the collective period style (instantiated in other aggregated artifacts) of the same epoch. The style is "horizonally" real in accord with our interpretive "prejudices" ((8.35), (8.36)). ("Prejudice," you remember, really means what is "prejudged" or "preformed" in terms of "judgment" before conscious judgment. The point is Gadamer's: a clever reading of Vorurteil .)
But, of course, if (10.12) be admitted, together with the impoverishing discrepancies assigned a moment ago to the "externalist" programs of analytic philosophy, then it follows at once that persons or selves cannot be straightforwardly identical with the members of Homo sapiens sapiens . (I have already affirmed this thesis with (9.13).) I have no wish to deny that we are both persons and members of Homo sapiens . I only say that saying that is profoundly equivocal. Since, for logical reasons, true identities are necessarily true, we need to make room for such possibilities as: (i) individual members of Homo sapiens that are not or not yet persons (fetuses, perhaps); (ii) nonhuman creatures that are persons (Martians, perhaps); and (iii) the conceptual possibility, supposing the pertinent reductionisms fail (as I believe they do), that persons cannot be "merely" numerically identical with any of the individuated members of Homo sapiens . That at least is what was anticipated in introducing the strenuous idiom of cultural emergence, incarnation, embodiment—and, more subtly, unicity and unity, and symbiosis.
I press examples from the artworld, I must say, because, like persons or selves, they too are "culturally emergent" and "embodied," without yet admitting that they are "minded" or have mental or psychological attributes. In a word, artworks
good or bad, or Nietzsche's Übermensch, or the God of the Christians according to Iris Murdoch).[10] This confirms, of course, that
(10.15) the mind/body problem
the culture/nature problem,
although, in the human context, the first cannot fail to implicate the second—or (asymmetrically), the second, the first. In fact, on the argument offered, it is not possible to solve the first without also solving the second. It should be clear, then, that
(10.16) the Intentional is not restricted to the mental, though it presupposes it;
and
(10.17) the mental is either inherently Intentional or Intentionally modeled ("anthropomorphized").
In the same spirit, I now add:
(10.18) the cultural or Intentional (or collective) competences that define persons or selves are acquired naturally .
By naturally, I mean only that new-born human offspring (Homo sapiens ) aptly and spontaneously internalize linguistic and lingual Lebensformen merely by living and maturing among the mature members of a society.[11] But to admit the ("existential") achievement marked by (10.18) is hardly to admit that the Intentional can be adequately analyzed in terms of the precultural aptitudes assigned Homo sapiens: the process of affecting the change required
Furthermore,
(10.19) the enabling Lebensformen of particular societies change continually with, and as a result of, their continual exercise.
Theorem (10.18) is an obvious empirical fact; by contrast, (10.19) is profoundly puzzling—perhaps the "template" for that family of doctrines developed by Hegel, Marx, Nietzsche, Gadamer, Foucault, and Bourdieu, and largely absent from analytic philosophy. What (10.19) signifies is the dynamic nature of a living "tradition": in which, that is, the aggregated use of a society's collective habits and practices changes those practices in a historically continuous way. (This is precisely what Gadamer means by a tradition and Bourdieu by a habitus.[12] I coopt these terms, therefore, although without subscribing to their authors' theories of them.)
It is important to bear in mind that, although our Intentional aptitudes are formed by internalizing the enabling practices of a tradition (to speak metonymically of a Lebensform, a tradition, a habitus, an episteme, a practice, an institution), there exists no tradition that we internalize. Discourse about traditions is entirely predicative. A tradition is (i) an Intentional structure, and (ii) a collective structure, (iii) nominalized over the culturally enabling powers of a parental generation with respect to its offspring, (iv) deployed historically, (v) entrenching prima facie norms of propriety or reflexive conformity, under historical change, and (vi) characterizable, heuristically, by the apt agents (persons) it enables to emerge, by way of rules or rulelike regularities. A perspicuous clue is offered by Wittgenstein in Philosophical Investigations (§ 202): " 'obeying a rule' is a practice." There is, however, no rule that is (or need be) obeyed. A rule, I should add, is (i) an Intentional uniformity, (ii) ascribed to a tradition or practice, (iii) imputed in either a realist or heuristic sense to (iv) the behavior, action, work, or thought of apt cultural agents, (v) possessing normative import (as of cultural propriety or aptness), and (vi) capable of extension under the conditions of evolving, open-ended societal life. In chess, the rules of play are genuinely constitutive and regulative of admissible chess moves—and are such as to form a closed system. In natural language usage, there are no actual, binding rules; nevertheless, apt discourse is rightly characterized in rulelike ways. (Wittgenstein notes that as well.) I should add: (vii) rules are adequated to a model of rationality (holistr ).
Every parental generation, let us say, is altered and affected by its role in rearing its own successors, and every successor generation differs from its parent in virtue of the changed Lebensformen it must internal-
ize. This is the generic lesson of what Gadamer calls Horizontverschmelzung ("fusion of horizons"). It signifies that thinking is historical, that the cultural world is a flux, that the metaphysics of Intentional entities is sui generis .
The key is this:
(10.20) paradigmatically, Intentional attributes are instantiated in the "intentional" life of persons and, hence, derivatively, in the artifacts they generate.
If we admit (10.20), we may infer:
(10.21) cultural entities do not exist, as such, except in a cultural "world,"
from which it follows directly that
(10.22) solipsism is incoherent.[13]
The point of introducing (10.22) so abruptly (although the matter has come up before) is to gain a march on the resolution of the puzzle of the relationship between the "mental" and the "cultural." There are, and can be, no solitary persons, except accidentally, as with Robinson Crusoe. (That was surely Marx's point, in the Grundrisse, where he ridicules the "Robinsonades" of contemporary theorists. It is also part of Peirce's inspiration for denying that pure "Firstness" ever obtains.) In short, selves and cultural worlds entail one another ((10.21)). I can put the point quite briefly: praxis (I should say) marks the conditional aspect of human powers that obtain as a consequence of (10.19); hence, it signifies that aspect in which theoretical as well as practical life (or reason) gains its competence and bearings in terms of the collective practices of its enabling (historical) culture. If I now add that reference and predication (as discussed in chapter 3) are praxical, you can begin to see how radical (10.19) may be judged to be. The use of the term "praxis " hearkens back to Marx's use and critique of Aristotle's Nicomachean Ethics in the Theses on Feuerbach . (I return later in this chapter to other aspects of the notion.)
The extraordinary importance of (10.22) lies with developing a reasonable policy for individuating numerically independent persons—in accord with the line of thought (offered in chapter 4) that links "exter-
nal" and "internal" relations between subjects and objects within a symbiotized space. Since Intentionality does not obtain in any merely physical world—it cannot be "adequated" to merely physical predicables if it cannot be analyzed or explained "reductively"—there is a danger of not being able to account for the reality of cultural worlds in a sufficiently robust way. I am trading, of course, on two themes that have already been secured: (i) the symbiosis of world and language; and (ii) the symbiosis of "external" and "internal" relations. We now find that, in the cultural world, a third form of symbiosis must be acknowledged, viz.:
(10.23) human selves are individuals only insofar as they effectively share the collective practices of a common Lebensform .
I recommend therefore that we hold that
(10.24) persons or selves are, by definition, culturally apt agents,
that is, agents capable of constative and other linguistic and lingual acts. (I shall return, in chapter 11, to the question of what an agent is.) I do not say the "natures" of human "agents" are nothing but their aggregated acts nominalized or assigned on the basis of such acts. (That would be close to the spirit of reductive behaviorism, except for the fact that Intentional predicables are anathema to thinkers like Skinner, although not, let it be said, to thinkers like Pavlov.)[14] My point here is that
(10.25) persons or selves are individual entities only insofar as they are aggregatively symbiotized relative to the collective Lebensformen they share.
Put more provocatively:
(10.26) the very existence of persons cannot be construed, either ontically or epistemically, in terms of a model of "external" relations that intrinsically lack Intentional attributes.
To admit theorems (10.24)–(10.26) is to admit at a stroke:
(10.27) the criteria for objective truth-claims in the human sciences cannot be modeled on whatever "naturalized" criteria may prove effective in the physical sciences;
for,
(10.28) persons cannot be mere spectators of themselves, or of one another, or of the cultural artifacts of their worlds, in anything like the way in which they function as external spectators of physical or natural or non-Intentional phenomena.
Alternatively put:
(10.29) persons are spectators of their cultural world only insofar as (as observers) they interpret the Intentional structures of whatever belongs to their world, that is, only insofar as they are competent "agents" in and of that same world;[15]
and
(10.30) there is no viable way of disjoining the (objective) perception and interpretation of cultural phenomena.
A whole raft of troublesome puzzles confronts us here. I content myself with two claims only and their joint import: one, that
(10.31) perception is linguistically modeled;
the other, that
(10.32) interpretation is perceptually entrenched.
You will remember that, in chapter 9, I argued that Intentional predicables are incarnate in non-Intentional predicables, and that (earlier in this chapter) I suggested that (as existent particulars) cultural entities are embodied in non-Intentional entities. (Embodiment and incarnation, I should say, entail a sui generis form of emergence, viz. cultural emergence, that is, a form of emergence that cannot be expressed in terms of the non-Intentional features and causal processes of the natural world.) I also indicated the need to confirm various forms of "adequation": in particular, (i) the "ontic" adequation of cultural entities and (their) Intentional attributes, as well as between physical or biological entities and their non-Intentional attributes, and (ii) the "epistemic" adequation of the conceptual and cognizing powers of selves and what they admit as the intelligible features of their world. Grant that much, and it will
be apparent, once you recall the third sort of symbiosis mentioned a moment ago (the aggregated sharing of collective Lebensformen ), that
(10.33) relative to perception and interpretation, Intentionality affects the "external" relations between cognizing subjects and cognized objects in different ways.
Certainly, however, before going further, we should explicitly affirm:
(10.34) persons or selves cannot be naturalized .
Theorem (10.34) is already entailed by (10.1), (10.2), (10.11), and (trivially) (9.13)—as well as (I remind you) by (9.15) and (9.19)—but it is also the compendious sense of the entire foregoing argument.
I draw your attention to the stunning fact that the distinctive aptitudes (cognitive and active) of persons are accounted for and legitimated in precisely the same sense in which the viability of reference and predication is legitimated, namely, in the real processes of collective life . What this shows is that if the inherent informality of reference and predication (reviewed in chapter 3) presupposes, for their characteristic epistemic success, the consensual processes of our Lebensformen, then the same is bound to be true regarding the cognitive and active competences of persons. This is the reason the culture/nature problem cannot be subsumed under the mind/body problem—the same reason the Chomskyan option is no more than a placeholder for a theory never supplied.
There is an enormous puzzle looming here. On the one hand, persons, as has been argued, are not identical with the members of Homo sapiens; on the other, they are the emergent artifacts of a collective culture, functionally identified as competent agents, in virtue of their sharing the collective practices of their enabling societies. There are other distinctive features of selves that I shall come to shortly, which bear on the relative indeterminacy and inconstancy of their "natures." These directly affect the precision with which we speak of persons as individual entities and specify the acts they commit. I don't deny that persons are individuals and that we treat them as such. But we are guided by different conventions in different milieux (law, medicine, biography), and even a certain informality, regarding the "unity" of persons. Nothing untoward follows from this, unless one insists on a greater precision than the domain will support—in fixing individuation, numerical identity, knowledge, effective action. In any case, examples drawn from the natural world
cannot rightly serve us here. (One sees this very clearly in epistemic contexts.)
Consider perception. I use the term perception in various wide and narrow senses to signify (i) the exercise of a cognitive capacity that is biologically grounded and culturally empowered (as in sensory perception, sensation, memory, feeling, introspection, general awareness, and the like), in virtue of which particular truth-claims may correctly express, report, or represent particular "perceptions," and (ii) the occasional exercise of a particular modality of (i) (as of vision, feeling by touch, fearing, being aware of believing, intending, and the like). I take senses (i) and (ii) to do general duty for nearly anything that may be said to be a distinct mode of cognizing. To this may be added further technical restrictions (iii) to what is admissible regarding (i) and (ii) (for instance, causal constraints on veridical vision or memory).
Furthermore, regarding the normal use of many of these modalities, we distinguish (iv) between a "nonpropositional" and a "propositional" ingredient, both of which may (perhaps, must, at times) be jointly instantiated in the paradigms of veridical perception.[16] Thus, in the matter of visual sight, exemplary cases admit the following paired sorts: "John sees a horse on the hill" and "John sees that there is a horse on the hill." These are essentially matched in veridical perception . John (veridically) sees a horse on the hill only if John sees that there is a horse on the hill and only if certain other conditions obtain "nonpropositionally" and noncognitively.
But there are also well-known asymmetries that hold in first- and third-person reports: for, if I merely say, "I see a horse on the hill," it may be correctly inferred that I believe I see that there is a horse on the hill, although I may not, relevantly, see that or "anything"; whereas if I say "John [veridically] sees a horse on the hill," one cannot then infer that John sees (or believes he sees) that there is a horse on the hill. The point of interest in this muddy problem is this:
(10.35) knowledge or cognition, which is at least lingual, if not explicitly constative, can be "adequated" only to persons or selves; otherwise, pertinent ascriptions are no more than "anthropomorphized."[17]
That is, theorem (10.35) follows from admitting that perception is propositionally structured . Hence,
(10.36) knowledge is an Intentional state,
which catches up the argument of chapter 5 against "naturalizing" strategies.
Sense (iv) of our tally about perception conforms, therefore, with what is affirmed in theorem (10.31), but that is hardly enough to ensure veridical perception . For, first of all, in "veridical perception" (in the case before us), there must be a horse on the hill to be perceived; for another, John must (we say) be in the right state to perceive the horse on the hill (veridically); and, for a third, it must be true that John then (nonpropositionally) perceives the horse on the hill and also perceives that there is a horse on the hill. Furthermore, for the second of these conditions to obtain, there must normally have been caused to occur a specifically non cognizing form of (nonpropositional) perception (v) in which veridical perception (iv) could rightly be incarnate (neurophysiologically, say). The clearest cases are afforded by sensory perception. Thus: to see a horse on the hill (in sense (iv)), John's eyes must have been suitably affected ("irradiated," says Quine) so that he appropriately sees (in sense (v)). It is certainly reasonable to suppose that a causal account may be given of the nonpropositional structure of perception (which, in addition, we suppose to be ingredient in perceiving in the cognitional way), but what should we mean by a causal account of events connected with the propositional structure of veridical perception or with the matching of the two?[18] (Certainly nothing nomologically manageable. I take the naturalistic maneuver to be a transparent dodge to avoid admitting legitimative questions.) Also, of course, these subtleties suggest a way of linking the biological and cultural aspects of perception and knowledge, but one that defeats our dreams of naturalized causal precision.
These busy qualifications explain why no adjustments relative to sense (v) can ensure "veridical perception" in sense (iv), and why naturalizing strategies are bound to fail. For, first, perceiving in sense (iv) can on no account be reduced to perceiving in sense (v); second, by veridical perception must be meant perception (in sense (iv)) that accords with JTB ("justified true belief") or a suitable alternative. I conclude that
(10.37) ascriptions of knowledge are honorific, relativized to legitimate convictions.
By honorific, I mean only that the ascriptions in question are never cognitively privileged or adequately characterized in psychological terms.
(To deny this is, in effect, to "naturalize" epistemology.) This is not to deny the importance of such ascriptions, but it does remind us that models of epistemic rationality ("rationalization") are themselves contingent artifacts of our lebensformlich practices ((5.14)–(5.15)).
Broadly speaking, an-archic strategies strenuously lower the philosophical importance of the role of truth and knowledge in the analysis of epistemic matters, resisting at one and the same time postmodernism and naturalism. An-archism refuses to dismiss second-order questions, but it is unwilling to treat them ("reductively") in first-order terms (causally or solipsistically, say) or in privileged (second-order) terms (as in externalist accounts). This is the point of theorem (10.37).
In short—to risk a leap that I cannot stop to support with the care it deserves—it now appears plausible to suppose that
(10.38) ascriptions of knowledge depend as much on presumptions of rationality as on evidence of the occurrence of the right causal sequences.
There are no ready ways of tracing causal sequences involving "propositional" perceivings: they already implicate the holism of our models of rationality. There is no way of overcoming the various sorts of indeterminacy affecting Intentional states and episodes, in terms of whatever precision we assign the description of non-Intentional states and episodes. And, for a third, individuated perceptions (beliefs, memories, sightings, even acts) are, more often than not, imputed rather than reliably discerned. In fact, there is here more than an inkling favoring a strong analogy between our policies in ascribing veridical perception and in ascribing moral validity to our actions and commitments. (I return to that theme in chapter 11. I am persuaded by the analogy.)
There are at least two further senses of perception worth mentioning briefly: for one, (vi) there is no developed sense in which some noncognitive form of the perceptual modality in sensation, in feeling pain or tickles or aches for instance, actually matches the fully cognitive form; for another, (vii) the pertinent modalities are distinctly intentional and either, as in (vi), no developed noncognitive sense obtains or, where it does, it does not "match" the cognitive sense in the required way. Thus, in Wittgenstein's reading of first-person "reports" of pain, there is no ready sense in which cognizing subjects can be said to know their pain, or that they are in pain, by perceptual means; they are said rather to avow (not to affirm constatively) what they feel. That which they
"avow" (that is, their sensations) cannot be independently perceived by others. Nevertheless, as in the dentist's chair or on the chiropractor's table, there are other causal and physical factors (rule-of-thumb correlations) that strengthen the sense in which avowals are cognitively constative after all.[19]
It is very much to the point, then, that although (felt) sensations are the "internal accusatives" of sensation-events, they are neither "independent" nor "intentional" (as Brentano realized).[20] Furthermore, as in Freud's account of the neuroses—in the "unconscious" fear of horses (said to be intentional, hence, mental or psychological)—the use of "intentional" does not play a role paralleling the noncognitive role of perception (v) that matches perception proper (iv).[21] Still, Freud's use of the intentional modalities (fearing, suspecting, and the like) is not altogether distant from ordinary ascriptions of belief, although it is restricted to explanatory inferences and assigned a twilight area a cut below conscious knowledge. (Even normal belief is often only inferentially ascribed and not directly affirmed or affirmable.)
The practice is also not very distant from the sense in which Chomsky claims we tacitly "know" the deep grammar of our natural language—not, however, in that way in which we normally make truth-claims about our language. (But Chomsky nowhere pursues the import of the "standard" sense of "know.") These considerations show once again the reasonableness of affirming (5.13) and (10.37). I mark our uneasiness about the "honorific" ascription of knowledge thus:
(10.39) knowledge is no more than a legitimated "status" publicly assigned "perceptually" incarnate states.
Hence, "knowledge" is a dummy state: neither a physical state nor an incarnate state—not a real state at all. Theorem (10.39) precludes all forms of cognitive privilege. By status, I merely mean, here, some normatively graded attribution (linked to certain standard interests—successful technology, for instance) that may or may not be construed in realist terms. Here, I remain agnostic about the matter. (In the following chapter, I raise the realist question directly.)
Wittgenstein's point, I should say, was valid against the externalist, but it is not in accord with his own notion of Lebensformen . The difficulty shows that Wittgenstein had not sufficiently thought out the import of his own doctrine. Probably, he was himself attracted to a form of externalism, although not to any as crude as Russell's or Moore's.
There is a matter of immense importance looming here, of which we have only a glimmer. It is this: since knowledge is Intentional ((10.36)) and since the Intentional is collective ((9.10)):
(10.40) knowledge "states" (perception, memory, sensation, nondeductive inference, and the like) signify as much the favorable consensual conditions in an enabling society as they signify either or both neurophysiological (or other physical or biological) conditions or psychological belief or conviction (or other subjective or private conditions).[22]
That is, no one knows anything except as a competent agent:
(10.41) cognition is a forensic competence.
Here, I take a leaf from Locke's theory of what it is to be a person. (Forensic means predicatively ascribable, said of states of competence and responsibility, in accord with consensually legitimated norms.) There's no question Wittgenstein had more than an inkling of (10.41). It is close to the master theme of the Investigations and On Certainty . But if he had seen it roundly, he should have seen that it would extend to reporting one's "private" pains as well. And if he realized that persons were themselves "constructed" in a lebensformlich way ((8.39), (10.23)–(10.29)), then he might have been led to a full acceptance of symbiosis. He was not so moved. At the very least, knowledge is a "state" paradigmatically assigned selves or subjects, open to realist ascription only if persons or selves may be construed in realist terms; and persons or selves may be said to exist only if their lebensformlich cultures may be said to be real. We see, therefore, the import of admitting the sui generis nature of cultural emergence: admit cultural emergence, and naturalizing and transcendental strategies fail at a stroke.
To accept all that leads now to a momentous finding. Although the assignment of various forms of competence and responsibility to persons (as cognitive and active agents) may be distributed to them as individuals, their competence cannot be assigned solely on grounds drawn from whatever is internal to their sub-cultural "nature"—physically, biologically, solipsistically, causally. For,
(10.42) agents are competent (and judged responsible) insofar as they instantiate the pertinent forensic (lebensformlich ) practices of their enabling society.
By responsible, I mean (as a term of art), judged (said of persons or selves) to be, or to have been, competent in the way of knowledge or action, or of being the apt agents of an enabling or surrogate society.
This shows very neatly the strong analogy between judgments of cognitive and moral agency. (I shall come to that in chapter 11.) What is more important for the moment is that we grasp how assigning a "nature" to persons or selves opposes any strong (individuative) disjunction between denumerable persons, contrary to what our individuative practices may recommend—particularly if we take physical bodies for our paradigms. That is,
(10.43) Intentional entities, having collective "natures," are individuals only as instantiative sites of collective attributes: in particular, persons are so individuated only as agents; they are individuated Intentionally and only as Intentional agents.
Hence,
(10.44) the objective description and explanation of the attributes of cultural entities (persons, artworks, words and sentences, and the like) cannot be disjoined from the description and explanation of the lebensformlich practices of an entire viable society.
This, for instance, explains at a stroke the validity of Putnam's notorious remark: " 'meanings' just ain't in the head! "[23] We individuate persons in a strong sense when we are attracted (with Williams, say) to the dictum "one person, one body", but we resist the strong analogy between individuating Intentional and non-Intentional entities when we feature the interpretable and collective—even horizontal—"nature" of cultural entities.
I can now also explain what I mean by speaking of perceptually independent things (physical objects, paradigmatically); I mean anything that (i) exists, (ii) lacks Intentional properties intrinsically, (iii) is in principle publicly perceivable in accord with exercising one or another of the modalities of sensory perception, and (iv) may be confirmed as "veridically" thus perceived. Anything meeting at least conditions (i)–(iv) is, I should say, perceived objectively . (This is the canonical sense favored by externalists: I have no objection to it.) I am also entirely prepared to extend these distinctions wherever needed, for instance, to account for the objective standing of theoretical entities: namely, those
that, in addition to meeting conditions (i) and (ii) but not (iii) and (iv), are (v) imperceptible (on some pertinent theory) and (vi) posited for the purpose of causally explaining what is perceivable in accord with (iii) and (iv).[24] It was in this sense that I had earlier introduced theorems (10.27)–(10.30). Other adjustments may well be needed, particularly regarding the perception of Intentionally qualified attributes and entities (as of paintings and persons). All this may be reconciled with Secondness.
It is easy to lose one's way here. The relevant distinctions form a thicket that the advocates of the causal theory of knowledge (Goldman, for instance) and the naturalizing of epistemology (Goldman again, since the causal theory is the principal strategy of the "naturalizing" school) tend to discount too quickly. My own strategy has been to show that knowing ("perceiving") is a salient power of persons or selves and, since it is inextricably bound to legitimative concerns, cannot be disjoined from the complexities of the Intentional world. Very simply put, this means that
(10.45) the "folk-theoretic" account of knowledge is conceptually insuperable.[25]
Theorem (10.45) affirms (9.25) more challengingly. The relevant argument has been greatly strengthened, for, now, to deny (10.45) is to claim to be able to show that the symbiosis of language and world and of "internal" and "external" relations within the terms of the first can be effectively retired. I cannot see now how that can be done. If the intended denial fails, then the "folk-theoretic" entails a third sort of symbiosis (mentioned before but not strenuously pursued): (iii) the symbiosis that holds between the individuated "integrity" of persons and their aggregated sharing of some collective Lebensform ((10.23)).
I know of no self-described physicalist who opposes the "folk-theoretic" (Churchland, for instance) who has directly addressed this puzzle. (I should perhaps add—recalling (9.25)—that, by a top-down methodology, I mean no more than this: that constative discourse cannot be "reduced" in any way not "adequated" to the resources of the folk-theoretic. I remind you once again that I have in mind the puzzles of reference and predication, which, as I say, seem to elude every conceptual strategy that is not "folk-theoretic."
There are two striking objections to the "reduction" of the folk-theoretic (ultimately, the first is a version of the second): (i) the "collec-
tive" cannot be reduced to the "solipsistic"; (ii) the "Intentional" cannot be reduced to the "non-Intentional." I put it to you that if these barriers cannot be breached, then the essential thread of the entire foregoing argument cannot be effectively opposed. That is a very considerable advance. But if you accept it, then, I claim, you cannot fail to go on to acknowledge that the objectivity proper to the human sciences cannot be "reduced" to the "objectivity" said to be "canonical" for the physical sciences. We are being pressed, therefore, to admit that the canonical ("externalist") picture of "objective" science cannot be right—or, better, cannot be enough even for its own work . The reason is plain: if all the sciences are human sciences ((9.21), (9.58)), and if the human sciences cannot be modeled solipsistically ((9.7), (9.59), (10.1)), then neither can the physical sciences! Q.E.D. The physical sciences are, I claim, abstracted from and abstracted within the terms of objectivity appropriate to the human sciences ((9.60)).
Grant that, and a decisive corner will have been turned. The key is this:
(10.46) the elimination or reduction of the Intentional entails the advocacy of the solipsistic.
Theorem (10.46) is the essential pivot on which the dominant programs of analytic philosophy (the canon ) founder. I can collect all of this now in a most strategic way:
(10.47) the normative function of epistemic appraisals cannot be reduced to, or replaced by, causal explanations of the psychological states judged to count as cognitions.
The first, of course, addresses what is incarnate, and the second what is incarnating; also, the first incorporates second-order concerns and the second does not.
We need to be clearer, however, about the nature of the "objectivity" appropriate to the cultural world. Again, the key is surprisingly straightforward:
(10.48) objectivity cannot but be consensual.[26]
Theorem (10.48) is a little startling, but entirely straightforward. By objectivity, I mean no more than that condition of our constative powers
in virtue of which our truth-claims can be legitimated as veridical: justifiably judged true (or otherwise suitably judged). Remember, we are not bound to bivalence. Of course, by the consensual, I mean (i) "entrenched" in the lebensformlich way, but (ii) not as such criterial. You have only to remember the discussion, in chapter 3, regarding reference and predication, to see that the matter was already implicitly decided by what we were there obliged to admit regarding our discursive resources. Under the various forms of symbiosis, consensual criteria governing the assignment of truth-values cannot but accord with our conjectures of what is "rational" in the way of validity and legitimation. Hence,
(10.49) objectivity, like truth, validity, legitimation, rationality, is a consensual artifact.
But what does the "consensual" entail? Try this:
(10.50) in an Intentional world, consensus is reflexive:
that is, (i) entails discerning what is collectively shared by aggregates of selves, (ii) in virtue of which each is empowered in pertinent cognitive ways. Compendiously:
(10.51) in an Intentional world, objectivity is consensual, reflexive, and interpretive.
Theorem (10.51) is no more than a summary. But from it follows:
(10.52) in an Intentional world, cognizer and cognized are the same;
that is, such that the "observation" of their "objective" attributes entails the interpreted perception of the lebensformlich structures they share . Hence, the observers of a cultural world are apt agents in and of that same world ((10.10), (10.29).) Alternatively put:
(10.53) persons or selves are, at once, cognizing agents and cognized cognizers.
Theorem (10.53) signifies, in effect, that insofar as persons are the apt individuals they are, their aptitude instantiates the incarnate collective practices that are their Lebensformen .
In short,
(10.54) in an Intentional world, there is no exit from the interpretive consensus in virtue of which such a world is constituted and perceived.
Alternatively:
(10.55) in an Intentional world, perception is inherently subject to the hermeneutic circle .[27]
Viewed more paradoxically:
(10.56) Intentional structures are real only insofar as they are cognizable by percipients whose own cognizing aptitudes are suitably qualified and conformably constituted.
Theorem (10.56) supplies the deepest version of that benign paradox known as the hermeneutic circle. I take it to define the "circle." Its puzzle cannot be resolved except within a lebensformlich world in which cognizer and cognized are constituted as such and "adequated" to each other, as in (10.52). I may now say that I mean, by the hermeneutic circle, that constraint on our cognitive powers such that: (i) our constative claims and perceptions are inherently informed by a more inclusive lebensformlich consensus; (ii) the "objective" determination of that consensus presupposes an interpretive consensus; (iii) every claim of interpretive consensus is an artifact of (i); and (iv) there is no escape from conditions (i)–(iii).
I trust it is clear that to admit the hermeneutic circle is to admit that "cultural emergence" must be sui generis, inexplicable in terms of whatever forms of emergence obtain in any world adequately characterized in non-Intentional terms. In effect, that means that, unlike biological or psychological emergence,
(10.57) cultural emergence is at once "natural" and legitimative.
Theorem (10.57) is an extraordinarily important finding. I can suggest its subtlety by reminding you that, in the psychiatric nosology, schizophrenia is often characterized as involving a deformation in the veridical perception of the objective world. But of course, on the account I have been developing, the "natural" phenomenon of schizophrenia cannot,
in principle, fail to be inseparable from its legitimative "status" vis-á-vis objective knowledge. The taxonomic manuals do not accommodate that fact—or, for that matter, its methodological import on the scientific pretensions of psychiatry. Another subtle consequence is this:
(10.58) since they are Intentional entities and since the Intentional is collective and horizonal, the integrity (the unity and unicity) of persons or selves is itself a function of our horizonal interpretation of their (and our own) careers.
Theorem (10.58) begins to open up dizzying possibilities, which, though still manageable, are clearly completely alien to canonical views of individuation and numerical identity. (Certainly, it is completely incompatible with externalist models of cognition.) Any surprise here is a consequence of the failure of analytic philosophy to address the culture/nature puzzle adequately. I suggest that it has been anticipated in recent views regarding the objectivity of literary criticism. Furthermore, both physical and cultural emergence are posited only at that level of "emergence" at which linguistically apt agents are first able to broach the question. It is in that sense that science is invariably "top-down."
I should perhaps add explicitly that, once you accept the general line of argument that leads to my reading of the hermeneutic circle, you cannot fail to grasp as well that it provides a similar treatment of praxis (in the general Marxist sense, shorn perhaps of the particular model of social causation the Marxists have favored historically.)[28] I take Foucault's notion of power/knowledge (pouvoir/savoir ), for instance, to be sufficiently "materialist" in this regard to be "Marxist." The idea is that the constructed standing of "knowledge" is a function of the historically contingent but practically effective scheme of institutionalized distinctions ("power") that some viable society has entrenched. The thesis is meant holistically, not criterially. Hence, it cannot be equated with the specific effects of deliberate political power, although it accommodates them. (A Lebensform, we may say, is a site of power, and knowledge is the artifactual status of legitimated exemplars within its space.)
The "hermeneutic circle," it's true, is normally not employed in the context of social causation, but it could be (and probably ought to be). In any case, by praxis, I mean (adding to what has been said): (i) the defeat of the conceptual basis for Aristotle's disjunction between "theoretical" and "practical" reason; (ii) the characterization of theory or theorizing (or language, for that matter) as a form of "practical" activ-
ity; and (iii) the interpretation of Lebensformen in terms of praxis and of praxis in terms of the enabling powers of Lebensformen. In short, the lebensformlich is the middle term between praxis and the hermeneutic circle.
Of course, objectivity, according to the "canon" (among the Positivists, the advocates of the unity of science program, those who would naturalize epistemology, those who would reduce the Intentional to the non-Intentional): (i) presupposes solipsism; (ii) admits a cognizable world that lacks Intentionality; (iii) denies theorems (10.51)–(10.55); and (iv) regards cognizer and cognized as epistemically independent of each other. Call this view of "objectivity," objectivitye (meaning that objectivity is confined to externalist relations), and the earlier view (in accord with (10.48)), objectivityi (meaning that objectivity accommodates internal as well as external relations). The two versions are clearly irreconcilable. That is the point of (10.45). (It signifies that "objectivitye " is a deformation of "objectivityi ")
Let me bring this discussion to a close with a final set of curious puzzles. Intentional properties are real only in the milieu of a viable culture ((10.56)), or only as the interpreted vestiges of a once-viable culture sustained by another that is still viable. Here and there, extraordinary conceptual bridges have been built between cultural worlds that no longer function "reflexively" and others that still do. (The Rosetta Stone is one of the most remarkable of these bridges.) Collective practices evolve through the effect of their actual exercise ((10.19)). I construe this to mean:
(10.59) cultural worlds are histories relative to which whatever exist within them are also histories.
On the strength of (10.59), we may now add:
(10.60) existent histories are Intentional entities embodied in non-Intentional entities,
just as Intentional properties (now, their Intentional properties) are incarnate in non-Intentional properties. This is the condition on which the "integrity" of "embodied" entities can be ensured—that is, the adequation of their "unity" and "unicity." I have already argued, in chapter 7, that individuated things need not have constant natures ((7.6)). It turns out, as a consequence, that, since the lebensformlich sources from which
the "natures" of embodied entities are drawn are themselves historicized ((10.19)), embodied entities cannot have constant natures ((10.58)).
Let us say, then, that
(10.61) all cultural entities are permeable.
By permeable, I mean that the "nature" of cultural entities is affected and altered as a direct consequence of changes in the Lebensformen in which they are formed and within which they "persist." Hence,
(10.62) selves are permeable because they have Intentional (consensually formed) "natures."
This explains the bearing of the "internal"/"external" symbiosis ((4.11)) and the open-ended nature of the hermeneutic circle on the methodology and "objectivity" of the sciences.
Furthermore,
(10.63) all cultural entities are porous because they are permeable.
By porous, I mean that cultural entities are capable of acquiring and losing Intentional properties as a function of the consensual practices on which their existence depends: that is, that their natures are intrinsically (Intentionally) collective. For example, Hamlet has acquired psychoanalytic import as a consequence of changes in the ethos of Western society—attributes that were not salient (or not even possible) in Shakespeare's time.[29] Similarly, the recent development of the feminist critique of "male-dominated" modes of life has entered into contemporary social criticism in such a way that, for instance, the "macho" life-style has been perceptually (interpretively, really ) altered. (Theorems like (10.62)–(10.63) oblige us to construe the "hermeneutic circle" in a historicized and relativistic way.)
This is not to say (as Dennett charges) that those who favor the "intentional" idiom hold that real things are altered as a result of mere redescription.[30] No, Dennett has misconstrued the problem. For one thing, non -Intentional entities are, on every view, completely unaffected by changes in mere description. (No respectable "intentionalist" has claimed otherwise.) It remains true, however, that, in intentional contexts ("opaque" contexts, as Quine calls them), it is not always clear whether one and the same entity is being referred to when descriptions
change. Here, I claim only that theorem (10.63) is not incompatible with what I have just said about non-Intentional entities.
Finally, given the historicized symbiosis of world and language (for example, along the lines Kuhn stresses, in admitting that Priestley and Lavoisier lived "in different worlds"), the "objective" concepts ("objectivee ") with which the non -Intentional world is described and explained are themselves the artifacts of our Lebensformen (of what is "objectivei "). In this sense,
(10.64) scientific objectivity is historicized.
Foucault was right. But I am also prepared to concede that
(10.65) whatever lacks Intentionality is "impermeable" and "nonporous."
The Hamlet example is profoundly instructive in other ways. (I count Hamlet as a metonym for the cultural world.) Over time, Hamlet has "acquired" a psychoanalytic import—as has, also, Oedipus Rex . But the Mona Lisa smile still possesses, vestigially, a trace of Verocchio's work.[31] Freud, by contrast, completely failed to see "in" Leonardo's Madonna and St. Anne the cultist meaning of rendering the two women equal in age: Freud imputed a sense to the painting that tradition has not sustained. Again, we have surely lost much of the semiotic import of the masks of West Africa, which have acquired traits born of being brought, in the museum world, into constant conjunction with mainstream European painting. And Dante's Commedia has obviously lost a great deal of the import of its Muslim exemplar (which may yet be restored). To the question, where are the Intentional properties of these works to be found and located, the best answer seems to be: in the works themselves (Intentionally complex artworks, mind you) as discerned in or imputed to them by Intentionally apt (even matched) agents. By imputed (to them), I mean only that the consensual traits (the "meanings," the "representational" properties and the like) they "have" are interpretively discerned in them. They "have" the changing natures they have in virtue of the rigor (naturans ) of such imputations; the imputed properties are objectively discernible in them, as a consequence (naturata ). In this sense, artworks (as well as selves) are "socially constructed" and continually reconstituted interpretively ((7.33), (9.23), (9.37), (9.38)).
This helps to explain the objectivityi of the cultural world—recalling,
always, that the seemingly rigorous sense of objectivitye in the physical world is itself dependent, epistemically and ontically, on just this objectivity.
I want to concede, however, that
(10.66) the individuation and predicated nature of cultural entities are labile.
By labile attributes, I mean those attributes in virtue of which embodied entities cannot be located physically, except by an interpretive convention, inasmuch as (i) embodiment
Where, for instance, does an action begin and end? or a war?[32] If, say, a particular act of flipping the light switch is "embodied" in a particular physical movement confined in some way to a certain (non-Intentional) causal event, then if (say, with Davidson) the action (or act ) of flipping the switch = the action of alerting a burglar (who happens to be in the house), then how may we confine that action to its proper "boundaries" in terms of its "unity" and "unicity": its "integrity," in short? Certainly not by appeal to whatever may be true of incarnating or embodying (non-Intentional) considerations. Furthermore, it is entirely possible that the practice of interpreting (a fortiori, individuating) actions according to the law may be very different from the practice of interpreting paintings. It is also possible that several different actions be embodied in the same physical movement. My schema is meant to accommodate all such puzzles. There is nothing comparable in our discourse about (mere) physical things.
I must press these gains a little further. I have already remarked that persons or selves are culturally apt "agents" ((10.24)). Let me add now that
(10.67) persons or selves are the only Intentionally apt agents there are,
in the sense that
(10.68) only persons or selves work or produce work.
(Remember, machines, prelinguistic infants, and animals may be "anthropomorphized" as "agents.") By work, I mean no more than: (i) acting in a way informed by our enabling Lebensformen (as in speaking or weaving a carpet); or (ii) the artifacts (or artifactual deeds ) produced (or effected) by work in accord with (i). Furthermore,
(10.69) an "act" or "action" or "piece of work" is done or performed only by individual agents,
even though what is done or produced will possess properties that, as Intentional, are intrinsically collective. This explains why that which is, predicatively, culturally significant is interpretively such.
The curious thing is that
(10.70) persons or selves are not only interpretable entities, they are also, uniquely, self-interpreting entities ("texts").
(The notion of a "text" I have already explored in chapter 4.) The important point here is that
(10.71) selves are the sole agents of history, though not the only causes of effective historical change.
Certainly,
(10.72) collective agents are fictions or "anthropomorphized" causes.
Braudel slights the role of particular human agents and favors instead the "agency" of large complexes of the Mediterranean world.[33] Certainly, for purposes of assigning "collective" responsibility, corporations are acknowledged in the law. By a corporation, I mean a collective agent (or person ): a fictitious or anthropomorphized agent (said to be capable of work and artifactual deeds and of assuming responsibility). But it strikes me that the convention of introducing "corporations" is an acknowledgement of the truth of (10.69). I add, therefore, following Locke:
(10.73) persons or selves are cognitively competent agents capable of work and of assuming responsibility.
The account is quite tidy now. Nevertheless, before closing this chapter, I must draw your attention to the import of (10.70). Selves, I say, act in ways in which, interpretively (whether consciously or not), they are the very agents by which, through their "permeability" and "porosity," they effectively alter their own "natures" (as well, a fortiori, as the "artifacts" of their own culture). Hence,
(10.74) selves are self-interpreting "texts," the continually reconstituted artifacts of their own reflexive agency.
I regard theorem (10.74) as capturing a good part of the strenuous but different doctrines espoused by Hegel, Marx, Nietzsche, Gadamer, and Foucault. It is a theme largely missing from analytic philosophy.
We are nearly at the end; a few more observations will round things out. Returning to a matter I treated too briefly before (but cannot do full justice to in this primer), I remind you that I introduced the notion of "embodied" in order to have it play an ontic role isomorphic with that intended for "incarnate" ((10.60)). The ramifications of (10.60) are extraordinarily rich, but for the sake of the most minimal clarity I add only these summary remarks:
(10.75) by definition, embodied entities necessarily possess incarnate properties;
and, as a consequence, drawing on the entire foregoing account,
(10.76) Intentional, intentional, mental, cultural, linguistic, lingual, interpretable, historical, artifactual, permeable, porous, labile, and similar properties are incarnate properties.
I should say here that by embodied I now understand (i) a complex "relationship" between numerically distinct individuals, (ii) one entirely different from that of instance/kind and member/class, in that only individuals may be thus related, (iii) an indissoluble relation as far as the existence of pertinent entities is concerned, (iv) such that their "incarnate" properties are "emergent" with respect to their "embodying" entities, and (v) open to being affirmed in a realist, heuristic, anthropomorphized, or fictitious sense. (By individuals, I mean persons or selves.
By individual things, I mean particulars, taken singly, usually in the sense of being apt for reference.)
There are a great many puzzles about the individuation and numerical identity of embodied entities. This is particularly clear when one thinks about plural (and differing) performances of the same Mozart sonata or plural utterings of the same word or sentence. I can just report that I have found all of these to yield perspicuously to the following distinction of art:
(10.77) all and only embodied entities are tokens-of-types.[34]
I hyphenate "tokens-of-types" deliberately. Particulars that are embodied particulars are taken instances of a type that is "itself" countable. Brahms wrote one "Fourth Symphony," we say, but "that" symphony has been performed innumerable times and in many different ways. We are not, however, obliged to think of types as abstract (existent) particulars of some sort that are (oddly) instantiated by (token) particulars. No, the "token" performances instantiate pertinent musical predicables, and we count those performances heuristically (nominalize them as events) as tokens-of-the-type for reasons of interest within our society (credits, permissions, royalties, and the like). The interesting thing is that the cloning and reincarnation of persons (regardless of the empirical facts) trade on the coherence of the same notion.
A further caveat is in order here:
(10.78) emergent phenomena need not be embodied or incarnate.
By an emergent order of reality, I may now say, more precisely than before, that I mean any array of empirical phenomena that (i) cannot be described or explained in terms of the descriptive and explanatory concepts deemed adequate for whatever more basic level or order of nature or reality the order or level in question is said to have emerged from, and (ii) is causally implicated and cognitively accessible in the same "world" in which the putatively more basic order or level is identified. For instance, the order of biological phenomena is said to have emerged from a more basic order (or "level") of inanimate physical things—without being incarnate in the sense here intended.
It is among questions of emergence that questions of adequation and reduction arise and are resolved. For example, Searle is prepared to make mental ascriptions of the brain, but the coherence and viability of
doing so depend on the demands of adequation and the prospects of reduction.[35] Two puzzles arise, which Searle does not satisfactorily examine: first, whether, if the "mental" can be ascribed to biological organisms (including Homo sapiens ), it follows that the mental can be ascribed to the brain; second, whether, if the first puzzle is favorably resolved, it follows that Intentional forms of the mental can also be ascribed to "parts" of the brain (neurophysiological processes, say).
You can see that there cannot be an end to these puzzles. The decisive point is that they all arise and are resolved within the terms of (10.48): the consensual nature of every sort of objectivity. It is curious, therefore, that conjecture about such puzzles as those just mentioned is so often insouciantly detached from the milieu of their very resolution. It is the irreducibility and sui generis features of the Intentional that, in the last analysis, are decisive.
Chapter XI
Values, Norms, and Agents
Some philosophers hold that the world has a determinate structure independent of our conception of it and that, although we ourselves belong to the world, the structure of our own lives is (similarly) independent of our normative reflections on that structure. As a consequence, values and norms are ontologically "queer" if they are thought to be discernible as "part of the fabric" of the real world. Hume was convinced that the use of the copula, "ought," signaled a violation of this important truth, but Hume did not extend his complaint to the entire range of moral values. In our time, Mackie has taken the drastic line, although also not unqualifiedly.[1]
It is difficult to formulate this radical thesis perspicuously, but it is the one that is currently opposed by so-called moral realists and their allies. The doctrine opposed is often called moral skepticism. In a fair sense, moral skepticism is a form of what I have earlier characterized as externalism. In my opinion, the "moral realists" and their opponents have, together, raised the crucial question regarding the grounds for the objectivity of normative practical discourse in law, in art, even in science, in addition to morality.[2] Not, of course, in the terms the skeptics would favor (and then attack), for the assumption on which they proceed runs completely contrary to the argument of this primer, but rather in the sense that, assuming the pertinence of the skeptics' challenge and the realists' rejoinder, the contest obliges us to explain how it is possible to reclaim objective moral values in a thoroughly symbiotized world that already implicates our interests. It is one thing, it seems, to recover a
sense of objective truth for science, under symbiosis; it is quite another to recover objective norms for conduct.
Now, this much is safe to say but completely unhelpful. The reason is not entirely clear. I believe the skeptic is right in some part of what he says, but wrong to draw the skeptical conclusion, and the moral realist is right in countering the skeptic, but wrong to think a mere minimal realism of values will ensure the moral objectivity wanted. Alternatively put, the argument between the skeptic and the realist is largely misplaced. The skeptic is mistaken about the force of his skepticism, and the realist is similarly mistaken about his realism, but both are right in countering the other's claim. They are arguing past one another. Properly perceived, the issue reveals a lacuna that the champions of objectivity and realism resist admitting—or actually deny—which they cannot fill to their entire satisfaction. There is, in fact, no single doctrine that can be rightly called "moral realism": at a certain point, it pays to abandon the contest as it is usually joined. There is a much deeper puzzle to reclaim. But the second is best reclaimed through a sense of the stalemate of the first.
My own thought is that the history of moral philosophy has misled itself with grander claims than it could ever support, but its detractors have usually failed to grasp the fact that their own line of attack subverts more than they would care to give up. It is true that the sciences require objective norms (of truth), but that, I suggest, is never sufficient (even if conceded) to legitimate (or to indicate how to legitimate) moral judgments in the sense usually championed. I daresay this, too, is a confusing thing to say.
The realism the "moral realists" oppose (in addition to opposing skepticism and cognitive privilege) is surely at least a certain extreme species of what Putnam has dubbed metaphysical realism, that is, the doctrine that (i) the real world is what it determinately is apart from our conception of it, and (ii) well-formed truth-claims are, as a consequence, true or false apart from whether we could ever confirm or know that they were.[3] On that intriguing (but preposterous) view, truth, however normative it may be, need not belong to the human world at all, although it is standardly supposed that values and norms cannot be coherently assigned anywhere but in a human world. This cannot be right if truth, whatever our theory may affirm, implicates some relation or other between language and world. In that sense, truth (or factuality) cannot be less "queer" than goodness, if goodness is queer when assigned a place among real things. Hence, both "metaphysical realism"
and "moral skepticism" must be defective theories—and for similar reasons.
In fact, if, by skepticism, we may understand doubts (open to legitimation) about the veridical correspondence between (internal) "representations" ("ideas" or representational beliefs) about the independent (external) world and that world, then, of course, if they are at all reasonable, such doubts will arise in the same ("externalist") way for truth-claims in general and for moral claims in particular. But the analogy fails to admit pertinent differences between "factual" judgments and "value" judgments. It fails to consider the differences between the (normative) legitimation of factual claims and the legitimation of value claims. It fails to do justice to the confused distinction between facts and values. And it fails to address the full propriety of skeptical doubts. My own view is that skepticism in general is (i) an essentially externalist program, and hence, (ii) completely obviated by admitting the various forms of symbiosis. The point is that there can be no principled skepticism if our cognizing aptitudes and the world's cognizable properties are similar for symbiotic reasons—or as they also are on constructivist grounds. But if that is conceded, then:
(11.1) there cannot be a principled distinction between the epistemic and legitimative prospects of science and morality.[4]
Nevertheless, (11.1) does not answer the question of the actual objectivity of science and morality or of the objective difference between them. Mackie, I suggest, fails to admit (11.1) but he sees something of the import of the dependent issues. The strategic point is this: once (11.1) is read in symbiotized terms, questions regarding the objective standing of moral judgments cannot be applied first to the prior space of the physical or natural world . There is no such prior world ((4.11)).
Putnam claims that truth has an inherent "regulative" or normative function, that "metaphysical realism" precludes such a function even with respect to the sciences it would champion; hence, that metaphysical realism is ultimately incoherent. He presses the argument in a way that (implicitly) precludes any of the forms of cognitive privilege either Aristotle or certain moral realists might be tempted to favor; he himself finds it necessary (and sufficient), in preserving the objectivity of both science and moral judgment, to fall back to a Kantian-like doctrine. The trouble is, the Kantian view, as we have already seen, is itself vulnerable to the charge of privilege (apriorism). Putnam has belatedly acknowl-
edged the bearing of this fact on his own program, but, in retreating to a less vulnerable post-Kantian symbiosis—in denying any principled distinction between the "subjective" and the "objective"—he has been too slow in admitting the obvious consequence for the "regulative" function of truth.
The usual moral realist is inclined to avoid the post-Kantian option (Platts, for example), and appears to advocate instead some sort of "externalist" view.[5] As a result, the (moral) realist lays himself open to the skeptic's charge. This cannot be said to hold of all moral realists, simply because some (Lovibond and Wiggins, for instance) would never accept the terms of debate the moral skeptic imposes on the contest—often (as with Mackie) a form of empiricism. And yet, even this more careful realist may still prove too sanguine. Casting the dispute in the moral realist's best terms will not vindicate a full-blown realism of moral norms. Nevertheless, the objectivity of moral judgment need not be utterly lost. There's a pretty paradox.
Dummett, with whom Putnam debated the regulative function of truth for several years, agrees that there are constraints on meaningful truth-claims—in particular, that a bona fide truth-claim must be "decidable" (that is, that, in making a proper claim, we have in mind a determinate, finitely completable plan here and now for determining whether it is true or false). Dummett, therefore, presses in the direction of opposing the principle of excluded middle, but he does not (once "decidability" is satisfied) oppose the principle of tertium non datur . In this sense, Dummett is not entirely candid. For, although he brings truth into the human world (in the intuitionist's way), he regards that about the world which makes truth relevantly determinate, to be not actually subject, thereupon, to any symbiosis. For Dummett, intuitionism is only an epistemic or methodological scruple. He reinstates tertium non datur at once, having settled the decidability issue, which is to say he favors enough of the metaphysical realist's view (and privilege) to settle for some version of externalism. But that should invite the "skeptic's" charge (now, regarding truth and decidability, if taken to be normative). Mackie is not drawn to this extreme, but he fails to provide for the normative role of truth.
I have allowed certain puzzles to collect here, but I need to introduce an economy. Moral realists, we now begin to see, are of at least two quite different sorts: (i) there are those who allow their theories to be defined as bare responses to moral skepticism and metaphysical realism; and (ii) there are those who insist on a realism of moral norms but not
merely by way of (i). Both are keen on avoiding the stigma of relying on some obvious form of privilege. For the moment, it will pay to remark that "moral realists" may be either externalists or advocates of symbiosis. By an accident of history, the term moral realism has come to be associated with theories of the first sort. For convenience, then, I shall, where needed, characterize the symbiotized option as cultural realism. I oppose the first and support the second.
Putnam calls himself an "internal realist " because he holds that any Kantian-like "realism" compatible with the "regulative" function of truth (Putnam's Grenzbegriff ) must (against Kuhn and Feyerabend, for instance) resist any slippage in the direction of post-Kantian symbiosis. (And yet, of course, Putnam has also persuaded himself to slip in just this way, for he now denies that there is a principled or fixed demarcation between the "subjective" and the "objective."[6] But if he holds to this, he must abandon his Grenzbegriff .)
The "moral realists" (among whom Wiggins is perhaps the best-known) do not feel bound—as least as far as the standing of moral claims is concerned—to offer an opinion on either Putnam's or Dummett's options; they hold instead that human life would be "meaningless" if moral values and norms were not "real" in some very strong sense. (They outflank Putnam's worry and Dummett's confidence.) But they neglect to explain how this can be—in ontic and epistemic terms. Putnam's claim is that the sciences would be impossible if truth did not have a realist function. But since to have such a function entails that truth is "normative," it would be incoherent (Putnam believes) to hold that the objective world (the world science discerns) precludes values and norms. In that sense, Putnam is, effectively, an ally of the moral realists, but of a sort they are not likely to welcome.
I shall, therefore, take moral realism (provisionally) to hold (i) that whatever the metaphysics of the real world may be, moral values and moral norms are, in some sense, "objectively" recovered but not directly discernible in its space, (ii) that, however truth-claims are rightly processed, moral claims are similarly open to objective confirmation, and (iii) that satisfying (i) and (ii) entails life's being "meaningful, " the sine qua non (according to the moral realists) of there being objective values in the first place. (Moral realism tends to be externalist, but that marks a serious defect in its argument.)
My sense is that "moral realists" would not be satisfied with (i)–(iii) alone, although Putnam might, since Putnam is (now) a willing post-Kantian. No interesting moral realist would fail to insist on a further
condition, viz.: (iv) that norms and values conformable with (i)–(iii) must be robust enough to support objective judgments regarding what is "categorically binding" or "prescriptive" or "overriding" or the like vis-à-vis human agents whose lives and actions are under review (Nagel's sensible but unsupported thesis).[7] Here, moral realism fails or falters badly, for the realist insists that condition (iv) must be granted (on pain of undercutting whatever makes life "meaningful" in the first place). Hence the moral realist (Wiggins, for instance) relies too much on defeating empiricism or relies too much on analogies between moral norms and norms of truth and meaning (Lovibond, for instance).
What the moral realist holds at all costs is this: first, that norms and values are on a metaphysical par with whatever realism is accorded natural phenomena; and, second, that that status is ontically independent of our individual interests and conceptions. The contemporary analytic moral realist ventures very little more, although, of course, more standard "realists" (Aristotle) offer instantly stronger (privileged) claims that the moral skeptic would rightly challenge. The moral realist insists (i) that the skeptic must be mistaken in attacking moral realism in such a way as to render other normative parts of our human world unacceptably baffling (for instance, regarding truth and linguistic intelligibility), (ii) that to make provision for epistemic success regarding the latter question already entails the moral realist's thesis, and (iii), that, more implicitly than not, resolving the issues raised by (i) and (ii) is hardly improved by yielding in the internal realist's way.[8]
I concede the validity of (i), although it is not narrowly relevant to the defense of the moral realist's position; hence, I doubt that any argument conformable with (ii) can be successfully mounted. Furthermore, the moral realist must still answer the skeptic's charge, although he is right to insist (iii) that the internal realist's thesis will not secure his claim. Finally, to support both (i) and (iii) is, I suggest, to argue at cross-purposes. The options before us make for much more than a local skirmish.
The trouble with internal realism cannot be confined to the issue before us: both Kant and his successor (the "internal realist") utterly fail to assess the decisive realist role of collective cultural life (our Lebensformen ). Without positing a "cultural realism"—which is effectively secured by embracing symbiosis—the moral skeptic cannot be answered, except by retreating to some form of privilege (externalist or constructivist). Accept symbiosis and the skeptic's objection instantly falls away. Accept internal realism and you cannot then legitimate (as Putnam has
always supposed he could) a "regulative" role for norms (truth or goodness) robust enough to secure a constructivist sense of objectivity as strong as what the externalist canon has always claimed—illicitly among unity of science buffs, Aristotelians, utilitarians, and the like. Putnam is not clear on the matter. Ironically, Kuhn, whom Putnam has regularly opposed, is much more explicit about the collective nature of the norms of truth regarding science. In any case, I hold (against Putnam) that moral realism cannot be adequately succored by the internal realist's treatment of truth in the sciences . And I hold (for instance, against Lovibond, who is a moral realist) that the realism of linguistic "norms" (communicative norms for natural languages) cannot succor moral realism. Both maneuvers are inapt for the purpose at hand, although reasonable enough in their own mangers. Still, truth and intelligibility are normative concerns: they do regulate in some way what we mean by "objectivity." It's just that moral norms have a different place in the scheme of things.
I recommend, therefore, that we begin (again) as tentatively as possible. Begin, then, with the following terminological distinctions:
(11.2) values are predicables for some order of things judged in accord with a given norm or norms;
and
(11.3) norms are exemplary values in a hierarchy of values, or principles or rules or regulative procedures for "grading" and "ranking" things—preeminently, choices, judgments, commitments, actions—pertinent to realizing such values.
By grading, I mean that form of evaluation in which certain value predicates ("good," "right," "true," for instance) are pertinently ascribed to whatever, singly, satisfies a given norm. By ranking, I mean a form of evaluation in which value predicates are ascribed in the ordinal and comparative sense, in accord with the degree to which different things satisfy a given norm ("better," "ought," "forbidden," for instance). Grading is a "threshold" matter; ranking, one of "ordering."[9]
Here, I suggest we take a note from Hegel's criticism of Kant; it will prove an economy:[10]
(11.4) values and norms are real (in the "cultural realist's" sense) in that they are sittlich: consensually entrenched in the historical prac-
tices of a lebensformlich world, and hence, straightforwardly confirmable in "objective" terms in first-order inquiry.
I cannot see how any account of moral values or any other values—cognitive norms for the natural sciences, for instance—can be "realist," unless something like theorem (11.4) obtains. Surely values have realist standing only in a human world, even if they are not predicatively restricted to the quality of human life. The sittlich vouchsafes that condition at the same time it collects the effective values of different societies. (I should say at once that I take "sittlich, " like "lebensformlich, " to function as an English adjective.)
I insist, however, that "externalist" legitimation entails illicit privilege ((2.5)–(2.7)) and that legitimations based on symbiosis alone cannot match what the externalist canon has always wanted. There's the difficulty. The result is, we must scale back our philosophical expectations. Norms and values have a realist standing in the sittlich world—in being consensually operative in lebensformlich ways—but they are not valid, for that reason alone, in any strong legitimative sense. We cannot guarantee that those same norms and values rightly serve as the "objective" (or "real") values to which we ought to subscribe (categorically or prescriptively). The skeptic insists on the caveat and the moral realist would be well advised to admit the weakness of his claim. For the moment, I urge only a sense of the puzzle.
Norms arise in a sittlich world, and they are forever tethered to that world. We suppose some moral values are "objectively" better than others, although they are all equally sittlich . The normative standing of truth offers a tempting parallel, but it is not an adequate examplar for legitimative strategies in moral matters. Also, insisting on the merely sittlich sources of moral objectivity trivializes the entire question. For there is then no argumentative ground for distinguishing between what is and what ought to be .[11] The sittlich does, however, obviate the question of whether life is "meaningful," for there is then no sense in which, collectively, life could possibly be "meaningless." (The moral realist's [Wiggins's] worry is ultimately negligible.)
Putnam is undoubtedly right to insist that truth is normative (since it is legitimative). But Putnam does not explain the link between the "regulative" function of truth and its sittlich standing in anything like the sense of (11.4). On the other hand, the moral skeptic cannot be entirely right, if (as is true) his objection against the moral realist also (inadver-
tently) counts against a realism regarding truth (under conditions that preclude privilege).
The fact is, the moral skeptic (most familiarly, Mackie) opposes the objectivity of moral norms on externalist grounds, that is, contrary to the doctrine of symbiosis ((4.11), (2.7)). Mackie is right to hold that, within those terms, a realist defense of moral norms must fail: externalism refuses to countenance the only grounds on which the question promisingly arises (symbiosis). But Mackie himself fails to grasp that the objectivity he favors also requires some sittlich consensus ((11.4)). The natural sciences cannot be altogether wertfrei (as Putnam rightly sees) if truth is to have a regulative function in ensuring objectivity.[12] Moral realists, however, would be wrong to suppose that correcting for that would be sufficient to provide a basis for the objectivity of moral norms going beyond the merely sittlich . That would be a complete non sequitur . I know of no moral realist (of the current analytic stripe) who has entirely mastered the force of this worry.
This also bears on Putnam's dilemma. As an "internal realist," Putnam is right to hold that the "regulative" function of truth can only be drawn out of a Kantian-like holism, but he nowhere explains how that is to be done robustly enough to save the objectivity of the natural sciences. (It cannot be done except on apriorist grounds.) Putnam seems to have persuaded himself earlier, however, that, having shown the incoherence of the "metaphysical realist's" account of truth, he was instantly entitled to claim an externalist's assurance regarding truth's regulative function—within his own internalist vision. But that proved impossible—incoherent, in fact—and he has now admitted it. The trick is to find an external, but not an externalist, formula suitably fitted to the sciences. Failing that, we cannot hope to resolve the moral realist's question convincingly. The puzzles of the philosophy of science and of moral philosophy are on a par metaphysically, even if objectivity cannot be secured in both domains in the same way ((11.1)). So Mackie's challenge cuts more deeply than Mackie may have realized.
The best Putnam could have offered would have been that truth can be (trivially) recovered as a mere sittlich value. The moral realist similarly fails to grasp that, in spite of the difference between the non-Intentional world of nature and the Intentional world of human practice, theorem (11.4) can recover truth and moral norms only in that sittlich way.
More is surely needed to secure the full ("objective") standing of moral norms (as well as truth). Moral norms are at least sittlich, but
apparently they must be more. Objective norms, we say, sit in judgment on our sittlich values. I should add, here, that I deliberately speak of "full moral objectivity" (or "full" epistemic objectivity) when I mean to distinguish sharply between the objectivity of moral (or epistemic) norms as sittlich and their supposed (full) objectivity functioning as "categorical," "overriding," "prescriptive," or in some similar (legitimative) way. The idea is that, in the familiar literature, moral norms are thought to be capable of confirming what, "all things considered," we ought to do in pertinent contexts, or what is "timelessly true." We are not generally tempted to conflate the sittlich forms of truth's regulative function with the legitimative grounds for its yielding full objectivity—unless that is indeed what Rorty is after—but we are tempted by the option in the moral sphere.[13]
In chapter 10, I explored what we should mean by objectivity . We now see that the distinctions offered there cannot resolve the puzzle before us. The reason is this: if the naturalizing strategy fails ((5.13)), then the objectivity of truth, moral norms, legitimative distinctions, and similar concerns cannot be secured in any causal (first-order) way. I suggest the following, therefore: by objectivity, let us now understand the normative (or, better, the legitimated ) "status" of our truth-claims about the world ((10.39)), regardless of the sector of inquiry to which they apply. In that way, we confirm that "objectivity" is a legitimative (a fortiori, a normative) notion (hence, that Putnam is right about that much, whereas Mackie is mistaken), but that the criteria of objectivity need not be (and may not be able to be) uniform for the whole of the real world. I have already distinguished between Intentional and non-Intentional reality. We must, therefore, adjust our criteria of objectivity to what is suited to the cultural world (in the consensual sense, that is, in which human agents are both observers and observeds) and to what is suited to the physical world (in the external, not the externalist, sense posited within the terms of the first). It's one thing to provide for the legitimation of truth in general, it's quite another to prepare this or that domain (physics or morality) for legitimated truth-claims. There's the point of the disanalogy between truth and goodness—and of realism in science and in morality.
The argument of chapters 9 and 10 confirms that the "external" conception is itself abstracted from the "consensual" ((9.57)–(9.60), (10.48)). Hence, the objectivity of the physical world is itself inextricably bound to the objectivity (and realism) of the cultural world .[14] I say only that that has not been sufficiently well grasped by metaphysical realists, internal realists, moral realists, or moral skeptics.
There is more that needs to be said, of course. But, without explaining what we should mean by a "realism" regarding values, we cannot fail to see that, since the professed objectives are either incompatible with theorem (11.4) or exaggerate its recuperative powers,
(11.5) "metaphysical realism," "moral realism," "moral skepticism," and "internal realism" are—all of them—either incoherent or defective.
Objectivity, then, is (i) the normative "status" of truth-claims about any part of the real world, (ii) epistemically construed in accord with selected criteria suited for grading particular claims, which criteria are (iii) themselves suitably legitimated. I hold that condition (iii) is either consensual or externalist and that the externalist option must fail ((2.7)–(2.9)). It would still be appropriate (admissible in a dependent first-order sense) to say (iv) that objectivity concerns the conceptual (external ) "fit" between what is constatively affirmed and what may be found in the world. It does not matter that, in addressing (iii) in terms of (iv), we operate under the constraint of the forms of symbiosis developed in chapter 10. For, on the argument, all forms of objectivity are thus constrained, and the question of specific criteria does not yet arise.
The important point lies elsewhere, in the fact that
(11.6) "objectivity " is a second-order predicate, in two senses: first, because it applies to cognitive "states" (or truth-claims) and not to any other referents in first-order discourse (unless dependently); second, because, in epistemic contexts, it applies only in accord with legitimative criteria.
Theorem (11.6) is important. It bears, of course, on the fortunes of naturalizing strategies. But, more to the point, it confirms that
(11.7) questions of legitimation, like questions of truth, validity, rationality, and the like, are, ineluctably, questions about normative values.
There you have the link between moral theory and epistemology: "true" and "good" (or "right") are normative predicates applied within a sittlich world.
Hence, recalling condition (iv),
(11.8) science itself is not viable if values and norms are not "realist" or "objective" in the "consensual" sense.
There you have the reductio of all "naturalizing" strategies, for there is no known way to naturalize legitimation ((1.4), (1.5)). For,
(11.9) even if, construed in realist terms, first-order values and norms may be naturalized, legitimative norms, also construed in realist terms (and in a way that bears on the status of the other), cannot be.
Putnam opposes the naturalizing strategy, because he admits the normative nature of truth. This is perhaps what persuades him that truth can perform a regulative function that ensures, in the long run, some "Peircean" approximation to the objective world (independent of "the opinions of you and me"). But if Putnam were to admit that legitimative norms (the legitimation of truth and moral values, for instance) were, within his own "internal realism," second-order norms, he would be obliged to concede that truth cannot have a regulative function that is not inextricably entrenched in our tacit and contingent lebensformlich practices ("external" but not "externalist"). Putnam is committed to a paradoxical doctrine: he needs an externalist criterion but he gives us only an internalist argument.[15]
Now, by the consensual conception (of objectivity), I understand no more than the feature of any theory that accords with theorem (11.6)—a fortiori, with the three sorts of symbiosis sketched in chapter 10: that is, regarding world and language, "internal" and "external" relations, and the collective and individual in Intentional worlds. By the externalist conception, I obviously now mean: any theory that accords with the externalist analogues of conditions (i)–(iv) of the tally given a moment ago—a fortiori, any theory that does not conform with the first and second forms of symbiosis just mentioned (should anyone suppose the third to be too quarrelsome). I take the "externalist" position to be flatly untenable. (The essential key is given in theorem (4.11).)
There remains an important distinction buried in theorem (11.6). To work out what is needed, I must return to the issues of chapter 4. My concern had been to encumber truth and legitimation symbiotically, so as to oppose archism and to deny all forms of cognitive privilege. But I had already noted, in chapter 4, that truth may play a normative and "regulative" (but not privileged) role with regard to truth-claims, and that "truth" and "truth-values" play quite different roles in constative
discourse. The key theorem (there) was (4.20), which remarked that truth and legitimation are second-order notions. (I had, in fact, introduced the distinction between the "internalist" [consensual] and the "externalist" [non-symbiotized] notions of truth, but I did not bring them to bear on the analysis of values and norms. I return to these issues now, to remedy the matter and to draw out the connection between epistemology and moral and other forms of value theory.)
Everything hangs on a small, but neglected, distinction. Let me make it explicit:
(11.10) truth is a second-order normative concept that has no first-order use at all (unless derivatively or if cognitive privilege obtains), whereas truth-values are second-order predicates used in both first- and second-order discourse, "regulated" by whatever theory of truth we adopt.
Theorem (11.10) conforms very nicely with the lesson of (1.5), regarding the mingling of first- and second-order discourse. Read predicatively, truth-values are normally restricted to sentences or utterances. Even so,
(11.11) in an Intentional world, sentences and other linguistic (and lingual) utterances are admitted to instantiate consensual objectivity.
I should say they were "objective" in (at least) a sittlich sense—but no more than that, on the argument before us. Furthermore, if (11.4) cannot be denied, then we have as yet no other large "realist" option ranging over truth-values but (11.11). (I do not deny that, in the natural sciences, there are all sorts of facilitating considerations bearing on the saliences of Secondness; see chapter 6.)
It will pay to recall that, earlier, I said that "exists" and "real" were "second-order" predicates. What I meant and what I explained in chapter 6—for instance in offering theorems (6.26)-(6.27)—was that
(11.12) if, in first-order discourse, "exists" and "real" are treated predicatively, then they can only function, there, as second-order predicates.
The point I had in mind was that certain anomalies obtain if, in first-order discourse, we do not segregate "first-" and "second-order" predicates. I mean, by that, no more than that second-order predicates (i)
explicitly possess "realist" import, (ii) must, therefore, be segregated from "first-order predicates," which do not, but (iii) must, when pertinently uttered, in uttering epistemic and ontic claims, still be legitimated. Otherwise, paradoxes like that regarding Kant's hundred Thaler will surely arise to embarrass us. The moral realist, I believe, is worried about the import of (11.12)—and rightly so. This leads to the important finding:
(11.13) qua sittlich,normative values may be no more than first-order predicates;qua legitimative,they are second-order predicates that may be uttered in either first- or second-order discourse, but must be segregated in first-order discourse.[16]
By segregated, I mean the effect of partitioning first- and second-order predicates so that second-order predicates are not thought to be "descriptive" of the world in any way ((6.25)-(6.26)). By first-order predicates, I mean no more than (i) predicates whose utterance (in constative contexts) invites questions of truth and falsity because they are descriptive of the actual world, but (ii) have, as such, no explicit "realist" import. This distinction among predicates does not affect in the least the pertinence of legitimative questions raised with regard to first-order discourse. (It explains why the disquotational theory of truth, even if it were true, could not be counted on to retire legitimative questions bearing on truth-claims.)
An important corollary of (11.13) holds that
(11.14) value predicates cannot be merely sittlich and, at the same time, function in a full legitimative (second-order) way.
(Theorem (11.14) puts the quarrel about moral skepticism and moral realism in a clear light.) The matter is tricky: something is still missing. What I mean to say, of course, is that a second-order predicate possesses (or conveys) "legitimative import" explicitly, that is, when merely uttered, although it cannot (as such) actually legitimate anything without satisfying the appropriate criteria of legitimation (of truth, rationality, knowledge, normative value, coherence, meaning, or the like). That is why it must be segregated. Put more doggedly:
(11.15) only the satisfaction of the appropriate second-order (epistemic) conditions governing the use of second-order predicates can ensure their actual, "full" legitimative function.
Trivially, my saying that "P " is true does not make "P " true, but it does raise the (explicit) legitimative question (inseparable from the confirmatory question) whether "P " is true ((1.5)). Still, in saying " 'P ' is true," I raise the legitimative question merely by uttering a second-order predicate ("true"); whereas in saying "This balloon is red," I do not explicitly do that . (I raise the legitimative question by speaking referentially perhaps.) The quibble is of little consequence in itself, but it is generated by the need to segregate first- and second-order predicates in first-order discourse, and it bears on a deeper matter affecting moral realism. ("True" is a second-order predicate. It obviously has sittlich sources, but it has no developed first-order standing despite a use in first-order discourse: e.g., "The jury found that what he said was true."[17] What we should take as the "full" legitimative functioning of "true," analogous to that of moral norms, depends on many things: the relation between theoretical and practical matters, the relation between Intentional and non-Intentional worlds, the accessibility of relatively undisputed criteria in different sectors of inquiry, and the like. Certainly, truth is thought to impose constraints on "rational" judgment and behavior and, hence, to have some function similar to the prescriptive function of moral norms. We say, for instance, that, on the evidence, a rational agent ought to take "P " to be true or ought to act conformably. These are not moral considerations.)
The legitimative question cannot be eluded. It is ineluctably posed by truth-claims, regardless of their grammar. But their grammar has proved so inconvenient at times that some have sought relief in such maneuvers as insisting on the disquotational theory of truth (Quine), the impossibility of treating "exists" as a predicate (Kant), and the "queerness" of value predicates in a realist world (Mackie). None of these maneuvers affects the issue at stake, simply because
(11.16) realism implicates objective norms;
and
(11.17) legitimative norms preclude "naturalizing" strategies.
Mackie fails to take into account anything like theorem (11.16)—effectively ruling out the pertinence of (11.17). Putnam accepts both (11.16) and (11.17)—but fails to attend to the constraints on objectivity that his own Kantian-like realism imposes. Both claims (Mackie's and Put-
nam's) are, on the argument being mounted, ultimately incoherent ((11.5)).
Given this much, I suggest that
(11.18) second-order predicables apt for first-order discourse include at least: epistemic attributes, attributes of rationality, ontic attributes, truth-values, other normative attributes conveying legitimative import, attributes of interpreted coherence, and the like.
Hence, given the evolving argument,
(11.19) the ineliminability of second-order predicables in first-order discourse entails the admission of an Intentional world.
But, then, of course,
(11.20) the admissibility of objective first-order claims entails the admissibility of objective second-order claims ((1.5)).
That, among other things, is what the futile effort to confirm the disquotational theory of truth was meant to derail. Admit (11.20) and you cannot, for purely formal reasons, then disallow the possibility of objective moral claims.
I spoke too quickly earlier of the sittlich nature of norms and values ((11.4)). I have of course registered a caveat in (11.14)–(11.15), and I still take (11.4) to be convincing. But what I said was too shallow and confusing. The right way to understand (11.4) (and the other theorems) is this:
(11.21) the sittlich is equivocal: the predicates it collects function either as first- or second-order predicates in first-order discourse; either way, the sittlich implicates legitimative questions—but, if so, then predicates may be legitimative in equivocal ways as well.
The point is this—in accord with theorem (11.15),
(11.22) sittlich predicates uttered as first-order predicates lack the legitimative import they have when uttered as second-order predicates; uttered as second-order predicates, their "legitimative" import still differs in important ways in the context of first- and second-order discourse.
Broadly speaking, what I suggest is this:
(11.23) sittlich legitimative predicates uttered in first-order discourse can be used (there) to "legitimate" only in a prima facie sense—perhaps then only trivially—whereas legitimative predicates uttered in second-order discourse are meant to legitimate in a critical sense and cannot (there) be merely sittlich.
This is what I had provisionally flagged earlier in speaking of the "full" legitimative function of moral norms (and, by analogy, of truth as an epistemic norm): prima facie legitimation
These are no more than verbal distinctions, to be sure, but they have a robust use. You must bear in mind that the moral realist needs to secure the epistemic use of critical "objectivity" (even if he supposes that that is no different from prima-facie "objectivity"). To say that Kwame has committed a serious breach of etiquette is, in a first-order sittlich sense, something a field anthropologist might affirm descriptively. To utter the same remark, speaking legitimatively in first-order discourse, is something a member of Kwame's tribe might confirm as a finding, in accord with custom (prima facie ). And to judge, "all things considered," that, whatever the custom, what Kwame did was or was not a serious moral affront is to judge (to presume to judge) in "full" second-order legitimative terms (critically). (It is this last possibility that moral realists must secure, which they cannot collect in the way they wish. )
This shows the reasonableness of theorem (11.21). It hardly justifies or legitimates the charge against Kwame. Doubtless, Hume had something of this sort in mind when he claimed that one could not derive an "ought" from an "is." We see now that Hume was mistaken (or, better, both right and wrong). He obviously did not consider the essential equivocation: of course we can derive an "ought" from an "is"—in the prima facie (legitimative) sense, and if in that sense, then possibly also in the critical sense .[19] There is no conceptual difficulty there, but there is a lacuna. If Jones is guilty of murder, then, surely, Jones did what, all things considered, he ought not to have done! Nothing could be plainer.
Nevertheless, a quite marvelous and surprising economy follows from applying (11.20) to the equivocation noted in (11.23), viz.:
(11.24) the (second-order ) legitimation of first-order sittlich values and norms cannot be entirely independent of the same sittlich sources.
That is,
(11.25) legitimation is at least sittlich.
The admission of theorem (11.25) is the only way, admitting legitimation, to avoid any ultimately hidden form of cognitive privilege. It is (11.25), precisely, that conveys the perspicuous power of a "genealogical" account of truth. There is a circularity that threatens here, of course, but it cannot be eluded, although it is easily resolved. It relates to what we have already encountered in Foucault's analysis of truth as both an artifact of history and an alethic predicate used "objectively" in claims about the history of truth itself ((8.36)). If you reflect on the import of (11.25)—in particular, if you construe (11.25) in the light of (1.1) and (1.4)–(1.5—you cannot fail to see the sense in which the realism accorded persons or selves and the world of human culture is the middle term mediating between the treatment of truth and the treatment of critical moral norms. In short:
(11.26) to admit legitimation is to admit cultural realism, and to admit cultural realism is to admit the pertinence of legitimative questions.
As a consequence, philosophy cannot but be "top-down" or "folk-theoretic." Even the plausibility of the eliminativist's argument cannot fail to be folk-theoretic!
I seize the opportunity, therefore, to make a small detour back to earlier questions. "Valid " (or validated ) and "legitimated " are, then (i) normative, (ii) sittlich, and (iii) profoundly equivocal second-order predicates, (iv) capable of servicing "objective" and "realist" claims (in the "consensual" sense). For, since
(11.27) we cannot exit from our Lebensformen,
we cannot deny that
(11.28) validity and legitimation are sittlich norms, in the double sense that: (i) they are grounded in sittlich sources, whether they func-
tion in first- or second-order discourse; and (ii) in first-order discourse, they function only as prima facie norms.
(Think of the disputed status of bans on abortion.) If, now, we bear in mind that
(11.29) the norms of truth are as sittlich as are our moral norms,
the missing theorem glides into easy view, namely:
(11.30) legitimation in the second-order sense must be more than sittlich, whereas legitimation in the first-order sense is sittlich only.
I have already christened the missing function, "critical."
We cannot, as Putnam correctly realized, give up the "regulative" function of truth. Yet, in accord with (11.14)–(11.15), we also cannot free that regulative function from its sittlich entrenchment. That is what Putnam neglects, what all externalists neglect (Davidson and Mackie, for instance). It remains true, nevertheless, that truth-values may be legitimated both prima facie and critically. Effectively, the sittlich is the lebensformlich viewed as normative[20] (i) in terms of the determinate historical practices of actual societies, (ii) confined to first-order norms and values, or (iii) carelessly or tendentiously, in confusing—with respect to second-order legitimative practices—"prima facie " and "critical" concerns.
Here, I must add a few clarificatory remarks. When, as in (11.28), I speak of validity and legitimation as grounded in the sittlich, I mean no more than that they are "entrenched" in our lebensformlich practices, or some habitus or tradition or episteme, although they presume to govern them in normative ways. I say that, with respect to truth, we usually introduce some (further) idiom to convey the specific sense of the prima facie: for instance, regarding what it is reasonable to "believe true" (as opposed to what is "true" tout court ).[21] That is, we are not apt to confuse the meanings of "believe true" and "true," although we may of course take what we believe true to be true. In moral matters, it is quite different: there, we are tempted to think that "objective in the critical sense" = "objective in the prima facie sense." That is the moral realist's pons. (Being "grounded," then, signifies no more than what is conveyed by (11.24). It has no criterial or privileged function.)
I take all this to confirm that
(11.31) qua critical,legitimation cannot be analytically derived from any first-order sittlich "legitimation."
Perhaps Hume had something like (11.31) in mind, but his own practice poses the question (despite his remarks about the opposed copulas). It is certainly not clear that Hegel would have sanctioned (or how he would have sanctioned) theorems like (11.30)–(11.31). My own argument insists that a satisfactory resolution must accord with theorem (11.8). The reason is simply this:
(11.32) truth-claims are interpretively adequated to our "constructed" worlds.
When, therefore, I say that the critical = the legitimative in the second-order (not the sittlich ) sense and yet is also sittlich ((11.30)), I am not contradicting myself. I signaled that fact by theorem (11.31). What I mean is this: in accord with (1.5) and (9.48),
(11.33) what, synchronically, is critical at time t is not, in that sense, merely sittlich then, but every critical claim or proposal, viewed diachronically, is interpretable, at t' later than t, as merely sittlich (or prima facie )—thus constituting further (sittlich ) data for some further critical claim at t' .
In short,
(11.34) viewed diachronically, the critical reverts to little more than the sittlich (or prima facie ), but, at every time t' at which that happens, a fresh critical conjecture at t' is not precluded.
In this double sense,
(11.35) the critical and the legitimative, as much with regard to truth as to moral norms, cannot entirely escape their sittlich sources and their sittlich fate.
It may be of help to say that
(11.36) critical discourse is inherently synchronic,[22]
that is, pertinently concerned with the reasoned coherence of the entire universe of discourse (at any time t ). (Accordingly, by diachronic, I mean no more than what, relative to questions of coherence, are the temporally ordered phases of synchronic discourse. Hence, "diachronic"
Roughly, what theorems (11.34)–(11.36) signify is that, in consulting the sittlich, efforts at critical legitimation (whether with regard to science or morality) consult not merely the history of our norms and values but also the history of our efforts at critical legitimation. Skeptics wonder how we should ever justify our efforts to validate critically the objectivity of our choice of norms and values. Part of the easy answer is surely this: whatever innovations we offer, we bring our proposals to bear on the ongoing practice of critical legitimation itself—on the fact that sittlich practice has already played a role in "yielding" prima facie legitimations as a result of accumulating presumptive efforts at critical legitimation. This is the instant benefit, of course, of shifting from externalism to cultural realism. Suddenly, our resources are enormously enlarged.
Clearly, on this reading, although I borrow the term from Kant, critical philosophies cannot yield anything like "synthetic a priori " (transcendental) truths. That is,
(11.37) critical discourse is as "horizonal" as first-order discourse, and for the same reasons ((8.31)–(8.32)).
The singular difficulty in arriving at objective critical norms—which I am inclined to think Mackie must have had in mind—is simply that
(11.38) there is no "regulative" intuition in practical matters and conduct as robust as that of Secondness with regard to natural events and causes.
Something like the truth of (11.38) has surely impressed all those who advocate "naturalizing" epistemology, moral matters, the world of human culture in general. (They misread "externalist" objectivity, how-
ever.) Fortunately, theorem (11.38) may be tempered: the picture is not unconditionally bleak.
Let me prepare the ground a little. Bear in mind that
(11.39) the sittlich is consensually real and incarnate.
I now introduce (abruptly) the troublesome notion of a fact . There is no easy sense in which it pays to insist that there are facts, that facts are real (or exist), or, for that matter, that there are no facts, that facts are fictions. The idea of a "fact" is generated by the deeper idea that what we utter constatively is true if and only if what, in the world (or better, what about the world), "corresponds" to what is thus uttered. That, of course, is the same intuition in virtue of which the disquotational theory of truth treats the predicate "true" as otiose. (I have aired the matter in chapter 4.)
If, now, we recover Aquinas's excellent formula—freely paraphrased as "truth is the adequation of propositions (which are constatively uttered) and facts (which are that about the world that propositions, correctly understood, "intend")—then, construing the formula in a sense left deliberatively uninterpreted epistemically and ontically (which is hardly Aquinas's intention), propositions and facts may be readily regarded as the matched referents of any discourse that would validate our truth-claims (whether in first- or second-order contexts, whether about normative or non-normative matters).[23] As we say, they are simply propositions, from the side of language, and facts, from the side of the world. Interpreted in accord with the entire argument of this primer, facts are facts, relative to the brute world parsed, or, viewed within the symbiosis of interpretations natura naturans and natura naturata, or, viewed within the hermeneutic circle. In a word,
(11.40) facts are artifacts.
They need not be fictions. In this sense, of course,
(11.41) the correspondence of facts and propositions is itself an ("external") epistemically legitimated artifact.
That is, "correspondence" need not invoke privilege. Also, construed in correspondentist or adequational terms, the concept of a "fact" is en-
tirely neutral as between an externalist conception of reality and any form of cultural realism ((11.40)–(11.41)).
As many have observed (notably, Austin and Strawson, in their well-known debate), propositions and facts cannot really be prised apart for epistemic purposes: facts are nothing but what true propositions intend; and propositions are true only if they intend what the relevant facts "are." For similar reasons, the correspondence theory of truth utterly fails wherever it is supposed to be criterial (as Wittgenstein realized in Tractatus ). Still, my reading of Aquinas's formula is completely unaffected by these considerations, simply because, being uninterpreted here, it is not (as undoubtedly it was meant to be) criterial of truth.
In the spirit of this benign reading, I offer two particularly strategic propositions about any world of values:
(11.42) facts and values form a mixed classification;[24]
and,
(11.43) the distinction, among sentences, between the supposed copulas "is" and "ought" forms a mixed classification.
I have already said, at the start of this chapter, that "values" are predicates or predicables ((11.2)). From what I have just added, "facts" are that (about the world) in virtue of which "propositions" take truth-values. Hence, without further ado, we may say that value judgmentsare such, in virtue only of uttering value predicates, that is, predicates apt for grading or ranking with respect to some norm or other ((11.2)); factual judgments, by contrast, are such, in virtue only of taking truth-values or truth-like values.
Of course, there is no reason why both conditions should not be jointly satisfied by the same token judgments. The entire habit of thinking that they cannot be has spawned the notorious "fact"/"value" distinction. That distinction, I say, rests on a palpable mistake: a confusion of grammatical categories. Hence, in accord with (11.42) and the realism of the sittlich, there can be no principled disjunction between factual and value judgments even on epistemic and ontic grounds. (By a judgment, I may say, I mean nothing more than a specific constative utterance intended to be treated as such, possibly following some pointed or strenuous deliberation.)
Regarding (11.43), it was Hume, as I say, who made such a great to-
do about doubtful inferences from "is" to "ought." But Hume mistakenly supposed he was addressing a formal distinction between two copulas, whereas he was actually promoting a certain doubtful form of empiricism (matched, in a sense, in Mackie). The inference from "is" to "ought" is not a mistake at all. It functions perfectly well in the sittlich sense (as we have seen), for example in the enthymematic argument sketched in the following: "Tom is a lifeguard at this beach. Tom is on duty now. Therefore, Tom ought to try to save that man calling for help from the water." Or, more familiarly: "John borrowed five dollars from me and promised to pay it back in a week. John owes me five dollars. Therefore, John ought to pay me back in a week's time." ("Borrowed" catches up the sense of (11.42)–(11.43).)
The idea that there is something logically wrong with these inferences is completely unconvincing ((11.22)–(11.26)). The point is that, in instances of the sort cited, ought is not a copula at all (or not merely) but a surrogate for a predicate of a ranking sort ("oughtful" perhaps). It would therefore not be amiss to paraphrase, "You ought to pay your debts on time" by "Your paying your debts on time is [deemed] 'oughtful' "—with whatever further accommodation may be needed to convey the imperative or prescriptive function usually intended. As a term of art—a predicate for rankings—whatever is "oughtful " (we may say) is to be preferred over pertinent options. The important thing to notice is that once we favor cultural realism over externalism, the entire attraction of "noncognitivist" options of the naturalizing sort—emotivism (Stevenson), prescriptivism (Hare), expressivism (Gibbard)—become both unnecessary and pointlessly pessimistic about moral objectivity.
What is important here is that the fact/value distinction may be resolved without invoking any ontic or epistemic considerations that adversely bear on the objectivity of value judgments—a fortiori, that adversely bear on the objectivity of token instances of prima facie or critical legitimation. The argument certainly does not strengthen the substantive needs of critical objectivity, but it removes a troublesome barrier. For, once you construe the Intentional world in realist terms ((9.30)–(9.31)), you cannot disallow the consensual objectivity of moral discourse . That's a very instructive leap. What remains is to fix the limits of rigor within which pertinent claims of truth and legitimation may be made.
One final nest of questions needs to be aired before bringing this primer to a reasonable close. Plainly, to say to Tom: "You ought to pay the loan back in a week's time, since you promised to do so" signifies that Tom is (in the normal course of things) capable (believed to be
capable) of doing so: capable of grasping his responsibility, capable of choosing to act responsibly on a reasoned choice.[25]
This raises, of course, the ancient question of human freedom . What I have to say on that score is short and sweet. I have already argued that persons or selves are "culturally apt agents," the exemplars of what it is to be a causally effective agent—entities capable of linguistic and lingual acts (or actions) effective in virtue of being embodied (and their attributes, incarnate) in physical movements (or physical attributes) ((10.12), (10.23)–(10.24)).
Now, it turns out that
(11.44) speech acts are the paradigms of human freedom and responsibility.
I cannot see how the admission of speech, particularly of constative and "illocutionary" acts (as Austin called them) can be acknowledged—with any sense of their complexity—without admitting, at the same time, that
(11.45) agents capable of speech in the usual contexts of natural-language discourse = free agents.
By free, I mean (i) capable of acting by choosing among real options and (ii) in accord with rational and evidentiary considerations; and, by responsible, I mean (i) free, and (ii) disposed to act conformably with rational and evidentiary considerations. These are very modest findings. But they do confirm that moral values and norms could not be other than grounded in our sittlich practices: human freedom is, congruently, a competence to choose among sittlich concerns. (It is not merely that, of course, but it cannot be far removed.)
In fact, when you think of matters this way, it stares you in the face that the paradigm of human freedom is nothing less than the ability to judge that a particular constative claim is true or false! That is an extraordinary economy and a confirmation of the intuitive force of the gathering argument of this primer. For how should we ever understand language in terms of linguistic (and lingual) acts if we did not admit our ability to choose or decide between the truth and falsity of a given claim? There could not possibly be a more compelling instance of human "freedom." Which shows, of course, that freedom is itself a legitimative, second-order consideration. Hence, for instance, if it makes sense to characterize persons or selves in terms of freedom or the power of
choice—which is what is entailed in construing persons as agents ((9.11), (11.45))—then whether human fetuses are persons is not a first-order question at all.
As far as I can see, the only serious ontic issue that might block the acknowledgment of human freedom is settled by admitting the following:
(11.46) it is not a necessary truth that, in the real world, the extension of the possible and the actual be the same.[26]
I take theorem (11.46) to be vindicated by (2.1). Beyond that, it is, as far as I can see, an entirely empirical question just what any particular agent is "free" to do. Theorem (11.46) entails the denial of strict determinism (that is, that the possible = the actual)—this denial is what freedom requires. Determinism is empirically unconvincing, in any case, among the physical sciences. But, in the absence of defeating the "folk-theoretic" vision (chapter 9), I content myself with merely remarking that
(11.47) the admission of a lebensformlich world entails the existance of free agents.
I take this to confirm that what I have already noted about the sittlich sources of the critical ((11.29)–(11.30)) is neither surprising nor a confession of the doubtful objectivity of moral norms. After all, on the argument I have been pressing, a strong analogy is bound to hold with respect to truth itself. By an agent, I may add, I mean no more than a person or self, an entity capable of being free and responsible with regard to thought and action.
Of course, saying no more than this tidies up what is needed for the small occasion before us. It says very little more about the human condition relative to our cognitive and active powers as persons. Take another moment to lay out the largest relevant options, however, both with respect to moral (and political) agency and with respect to our cognitive competence. With very few exceptions, the philosophical tradition from the ancients to Kant treats human selves or subjects merely adequationally . By an adequational theory of the self, I mean a theory that (i) admits objective knowledge about the world, (ii) with or without admitting objective prescriptions for our conduct in the world, (iii) as directly accessible to human beings; a theory that (iv) meets constraints (i)–(iii) by
imputing to human nature only what, antecedently, is abstractly required for that, (v) without interpretive tertia . Among moral philosophers,[27] Rawls is (by his own admission) an "adequational" theorist—both in A Theory of Justice and Political Liberalism . He remarks, in the latter, that his theory requires no more than a "sparse" sense of human selves or agents, whereas the usual criticism of Rawls—or of Hume or Mill or Kant (on Hegel's argument), or Habermas—is precisely that none of them attends to the complexity or historicity of the human condition within the terms of which objectivity is first gained.
To acknowledge that the human self is a culturally emergent artifact ((10.21)), an entity that has a history rather than a nature ((10.74)), and that cognizing subject and cognized object are symbiotized, is to favor instead an existential account of the self. By an existential theory of the self, I mean (once again) no more than a theory: (i) that includes at least the counterpart of an adequational theory, but (ii) construes the competences admitted in such a theory to be formed historically in accord with some enabling Lebensform, (iii) which then affects all questions of objectivity. This, of course, is an enormously important matter that I cannot do full justice to here. For my present purpose, it is enough to note that moral realists and moral skeptics are typically adequational, not existential, theorists, and, as a consequence, they generate insuperable paradoxes (Mackie, notoriously) regarding a realist reading of moral legitimation and knowledge in general. Only an existential account fully committed to symbiosis can resolve the legitimative question favorably, once privilege is denied. This marks the strategic importance of Hegel's distinction between Moralität and Sittlichkeit (and its analogue in factual matters). (But the resolution is far from automatic.)
I come, finally, to the matter of moral norms viewed in the second-order ("critical") sense. In the philosophical literature, Kant's conception of the Categorical Imperative is the obvious exemplar. But Kant's view is altogether too extravagant, since it involves (as it does) invoking (necessarily) the use of the concept of a noumenal world, the presumption of an invariant form of reason, and the absence of any explicit methodological presumption in favor of genuinely sittlich constraints. What I offer instead along related lines is the following: although theorem (11.31) is plainly true,
(11.48) it is prima facie "good" or "oughtful" (in the legitimative sense) to ensure our first-order sittlich values.
Furthermore,
(11.49) it is prima facie "good" or "oughtful" (in the legitimative sense) to meet our first-order prudential needs .
By prima facie, I now mean (i) an adverbial qualification of any attempt to legitimate a judgment, (ii) thought to be "provisional," (iii) restricted to sittlich considerations alone, (iv) which may be overridden for higher "critical" reasons. By prudential needs, I understand whatever values are (i) reasonably construed (empirically) as minimally necessary for a sittlich morality, (ii) grounded in our biological condition as members of Homo sapiens, or (iii) construed as instrumental to, or a functional part of, our sittlich values, or (iv) construed in terms of survival—individual, aggregated, societal, species-wide. (Note: prudential needs are normative but need not be moral.) There are undoubtedly other arguments of this kind regarding prima facie legitimations. I regard that as all to the good. For what they show is that second-order moral questions are not entirely without resource (open to being favorably compared with our intuitions about Secondness). For:
(11.50) any would-be prima facie legitimation of our sittlich values and norms by which aggregates of individual agents are willing to live, and do live, is, trivially, legitimate—prima facie .
I an entirely prepared to admit that theorems (11.48)–(11.49) are logically trivial although not at all trivial in practical terms. They are trivial in the plain sense that sittlich values are normative (relevantly, action-guiding ); they are not trivial in practical terms, because they require a form of practical consistency, the pertinent congruity between judgment rendered and responsible action. Both considerations implicate forms of rationality, but not the same form. Those who believe (like Anscombe and Hare) that "practical consistency" always requires (when valid) an agent's acting "straightway" (as Aristotle says) have failed to notice that (even on their own terms) they must first secure at least the critical legitimation of the norms they invoke in relevant contexts.[28] (They nowhere do that.) The confusion or conflating of the two (forms of legitimation) I call ideological. Still, there is a great gain here. For, on the argument, there can never be a dearth of (sittlich ) moral practices in terms of which to order one's life responsibly and in a "meaningful" way. That relieves "critical" speculation of the need to ensure a com-
pletely inclusive system of moral rules: it is, after all, entirely possible that not every sittlich need can be convincingly construed in the most exacting "critical" terms.
I must add (what Aeschylus and Sophocles make so much of):
(11.51) in moral matters, the mutual compatibility or incompatibility of different sittlich values and norms—a fortiori, regarding prima facie legitimation—is, as such, a matter of prima facie indifference.[29]
Theorems (11.50)–(11.51) are a little startling but hardly negligible. They do not signify that "anything goes," but they are hospitable to the widest range of divergent convictions and are clearly prepared for sittlich conflicts.
The question is: What do we expect from our moral theories? My own answer is a slim one: a modicum of humanity, an openness to conflicting convictions, an admission that there are no privileged norms, and a willingness to bear witness within the routines of ordinary life to whatever one regards as the sorry plight, the injustice, the inhumanity suffered by others. Very much more than that I believe is futile, philosophically, although it is also hardly enough politically and morally. To put more in place, one must presume to have greater moral resources than we can possibly justify.[30]
Beyond that, I have only three things more to say. First of all,
(11.52) critical norms for moral legitimation are socially constructed from, but cannot be discovered in, or analytically derived from, sittlich materials ((11.30)).
Mackie is surely right in this, although not for his reasons. Secondly, granting theorem (11.52),
(11.53) if there are any salient, relatively convincing moral norms in the critical sense, then the perception of summum malum may be made to yield a moral norm (or norms) as reasonable, in that sense, as any.
I mean, by summum malum, whatever (i) is spontaneously and widely perceived, (ii) notably across divergent cultures, (iii) to be a condition of suffering moving toward the limits of human tolerance, (iv) among whole peoples, large populations, well-defined societies, over extended
periods of time, (v) often among smaller groups and individuals perceived within similar terms, or (vi) to be forms of pain, deprivation, and the like that bear decisively on the survival of whole societies or the satisfaction of what are judged to be their prudential needs, as in starvation, torture, disease, natural disaster, war, genocide, willful murder, prolonged and extreme discrimination, and the like, (vii) usually in accord with sittlich convictions, but (viii) not, as such, a moral matter .[31]
I insist here that: first, the perception of summum malum is not and need not be of an invariant or unconditionally valid or even relatively constant sort; second, it is not as such offered as a moral principle or ineluctable moral value; third, judgments in its name are interpretively variable, informal, possibly conflicting, not always reliable; fourth, judgments of summum malum are factual judgments and, if normative (for instance, medically informed), not moral judgments as such; and fifth, in the moral commitment of individual careers, the admission of summum malum may not and need not play a central moral role and may, even if admitted, be overridden by other considerations that cannot, merely as such, be shown to be critically illegitimate. For example, the recent famine in Somalia has attracted a remarkably widespread perception of suffering—summum malum —both among aggregates of people and corporate societies, such that many have urged the need for immediate relief—in terms of various moral visions. But many have also felt no sense of moral obligation to relieve such famine, even when legitimating their own commitments.
The perception of summum malum is, I think, the exercise of a salient (perhaps a profoundly animal) disposition, not a moral intuition as such but a perception capable of spawning and inspiring (yielding, as I say) reasoned moral judgments and moral norms grounded in the biological sensibilities of the species, incarnating various sittlich values but detached from them (in any legitimative sense), and capable of overriding (when incorporated in a critical moral vision) mere sittlich moralities. It is a fortunate form of sensibility addressing what many regard (possibly under its influence) as among the most salient of moral concerns.
The pretty thing about this use of summum malum is that it eschews the greatest exertions of moral realism altogether. That is, it builds on the sittlich realism of morality—not enough for a critical morality (which the moral realist dearly wants)—and it proceeds parasitically (or dialectically) on the strength of whatever others in the society pretend is or can be legitimated in critical terms (for instance, "ideologically"). By parasitically, I mean that it seeks only a conditional legitimation:
that, given the critical or realist presumptions of any complex society regarding its own moral norms (ideologically), it is entirely possible to confirm that a counterpart morality (a "second-best " morality) based on summum malum and addressed to the sittlich values that prevail will, projected from these non-moral grounds, prove as reasonable as, or more reasonable than, the morality ideologically invoked.
If you grant this much, then a parallel argument can be mounted as well in terms of what may be termed minimum bonum: that is, normative, empirical, but nonmoral conjectures about what any human infant requires for a tolerable life, relatively free of pain and want and insecurity and the stunting of various forms of development, and the like. Again, I say, there is a considerable convergence here (of a nonmoral sort) among the peoples of the earth, relativized, to be sure, to the perception of technology and history and ideology. But if there is no noticeable convergence with respect to minimum bonum and summum malum (or similar dispositions), then, I fear, there can be no plausible sense in which a critical morality could ever be said to be objective (objectively "yielded"). Furthermore, of course, the doctrine of human rights is a perfectly plausible critical moral (political, economic, etc.) conjecture about proposed "rational" norms of societal life, (i) capable of being effected, (ii) dialectically convincing in terms of bringing minimum bonum and summum malum to bear on sittlich practices and actual ideological convictions, and (iii) not merely utopian but, within practical terms, able to support (in real-time terms) progress toward "indicative" universality. (Historically, the doctrine of human rights has been defended in archic terms—as natural rights . But that is an extravagance. I hold no brief for such a thesis.)[32]
The advantage of a "conditional" legitimation lies with its conceptual economy, the acknowledgement of the difficulty of strong realist claims, its generosity and openness to divergent views, its tolerance of conflict, its refusal to construe morality as akin to a scientific system, and, most important, its concession that morality need not be, critically, either categorically binding or comprehensive with regard to the entire range of human practice. By categorically binding, I mean that logical or quasi-logical feature of moral judgment by which it is alleged (as among the prescriptivists) that if a critical moral judgment is valid, then, for any possible alternative options within the pertinent space of action, to acknowledge that it is critically valid and to intend not to conform (although one is able to conform) to what it enjoins is to conduct oneself in an immoral, morally deficient, irresponsible, or irrational way. By
comprehensive, I mean only that, on the basis of some proposed morality, nothing in the way of commitment or action would not find its systematic place at the bar of objective validation and legitimation. I take both claims to be altogether uncompelling on the evidence I have been collecting. (Here, "comprehensive" = "all things considered," sufficient to determine what is "full" in the fullest possible critical sense.)
Roughly, my thought is this: corporate societies tend to insist on loyalty to first-order sittlich values more or less already in place, as well as to norms of legitimation that, although partly "critical," are defended partly on prima facie grounds (with whatever ideological skills may be invoked). In this sense, societies "act" corporately to bring the conduct of their aggregated members into some relatively "comprehensive" sittlich order. (So there is no danger of a dearth of "comprehensive" and "categorical" moralities.) Individual agents, however, often have no such overriding needs, and they oppose these tendencies on their own putatively critical grounds. They may have quite idiosyncratic visions—for instance, as did Nietzsche.[33] But, insofar as we may expect a measure of "critical" congruity between the collective morality of a entire society and the separate moral visions of individual selves, theorems like (11.48)–(11.49) are probably close to the core of the prevailing moralities and are reasonably realistic if offered as constraints on flights of moral fancy. In short, the account being offered presupposes the entrenchment of an effective sittlich order. I do not legitimate that order in the critical sense, but I acknowledge that, "corporately," it will act to compel all those within its space to believe, or to act in accord with, some variant of this dictum: prima facie legitimation = critical legitimation. In short, it will function ideologically . Moralities based on summum malum or minimum bonum may, then, in context, compete dialectically (I claim) with any such ideologically grounded moralities (as "second-best"). That is where the economies of critical reason play their best part. They need not pretend to any cognitive privilege, because they are parasitic, and they need not subscribe to the rigors of a bivalent logic, since they are not advocated in any sense stronger than the cultural realist's terms.
What is important, here, is that legitimative disputes about moral values and norms tend—however strongly committed to second-order questions—to be profoundly grounded in sittlich ways. Indeed, once the
argument against realisms that veer off in the direction of metaphysical realism, moral realism, and the argument that tempers moral skepticism are in place, it is difficult to see how a critical morality can fail to be substantially grounded in sittlich terms, or fail to be open to "parasitic" strategies. That finding will be strengthened if one also accepts additional limitations (second-order but not moral) on plausible candidate visions, such as the following:
(11.54) there are no convincing critical grounds for legitimating any uniquely valid or preeminent moral vision;
(11.55) there are no convincing grounds for construing the incompatibility or conflict of sittlich or prima facie values and norms, or of critical systems grounded in such values and norms, as signifying a failure of moral vision;
(11.56) there are no convincing grounds for legitimating any principle or principles (if not merely vacuous or tautological) as morally indispensable, essential, or necessary to any valid critical vision.
I take theorems (11.54)–(11.56) to signify that
(11.57) constative claims in the moral world favor (but not morally) a relativistic logic.
To insist on bivalence in moral matters presumes an access to objective moral sources that our sittlich world cannot convincingly sustain. In short,
(11.58) to confine critical legitimation within the terms of a bivalent logic (moral realism ) is to confine morality (and other normative practical concerns) ideologically.
Simply put, to admit the ineliminably sittlich character of morality at every level of rational review, to admit the loss of cognitive privilege, to admit the conflict among sittlich norms and values through the whole of human history, to admit the conflict of critical moralities, is to admit as well that
(11.59) there are no convincing grounds for insisting on a determinate, necessary, unique, practical or pragmatic or rational or moral connection between responsible moral judgment and responsible moral action :
first, because one may not believe, legitimatively, that (contrary to theorists like Hare and Anscombe) moral judgment, being practical (that is, action-guiding on the basis of the best reasons one can muster), must also therefore be prescriptive, or "categorically binding" (that is, such that, admitting the validity of a moral judgment that ranks one's own options, when the appropriate circumstances obtain and when one perceives them to obtain pertinently, one "straightway acts"—for rational or moral reasons or both); and, second, because one may not believe that, in that sense, morality "comprehensively" overrides (that is, takes precedence over) all other practical commitments.
I confess I am one of those who subscribes to (11.59) for the reasons given. At any rate, I see no way of demonstrating its incoherence or ineligibility. Hence, to end this account by offering the third comment I promised, let me say,
(11.60) there are no convincing grounds for disallowing, as contrary to reason, the view that, beyond morality, we function as civilized witnesses of the human condition; that, grasping the inherent limitations of moral inquiry and judgment and commitment, and however we may act morally, we may also act, overridingly—as rationally as any—merely by preserving the public memory of our lebensformlich history.
Part of any such effort lies with preserving the continuous public memory of perceived crimes against mankind, the record of mankind's suffering (summum malum, construed morally), the failure to satisfy minimum bonum, and the violation of conjectured human rights. But, more than that, it is a way of sharing, without let, the history of our global home.
The point is this. To be a civilized witness is to believe (i) that we may be able to change "human nature" in accord with our critical convictions, and (ii) that, by a division of labor, we may alter and enlarge our understanding of the entire world, so that, as a result, (iii) our sittlich resources may begin to incorporate what is critically favored in other societies and other worlds. There is no special wisdom in this. (And there
is obviously a great risk.) But it is a vision by adhering to which summum malum may conceivably be diminished and minimum bonum enhanced.
It is certainly not mere quietism, and it is not a merely moral matter. It is more a matter of coming to realize that
(11.61) a critical morality is a "rationalization" of a practical or existential vision that: (i) is, on legitimative grounds, rationally unable to invalidate all divergent, incompatible, even actively conflicting visions on the part of others; and (ii) is rationally unable to override effectively the prima facie norms of all others.
I should say that, by practical, I mean that aspect of a human agent's life in virtue of which judgment and action may be coherently "uttered" in accord with some critical or prima facie norm of rational behavior. By uttered, I now mean, by explicit extension from what has been said of speech acts, the instantiation, in the sittlich world, of effective lingual acts as well. By existential, I mean no more than the "practical"—construed horizontally by actual agents. (By "meaningful, " I add again [in agreement with Wiggins's usage, although not with his doctrine], that I mean the "existential" when actually sufficient to motivate the "practical" commitments of an entire human career.)
Finally, reflecting on certain alleged philosophical desiderata, I should say that the entire foregoing argument precludes any critical universalism —in any practical, existential, or meaningful sense: meaning by that, that (against philosophers like Rawls or Hare or Habermas)[34]
(11.62) practical judgment proceeds primarily analogically and by exemplar rather than by formal rule or algorithm.
If you grant (11.61) and (11.62), you cannot (I think) deny that
(11.63) if summum malum "yields" a critical morality as good as any (11.49), or minimum bonum, then, under its terms, no society palpably suffering summum malum or failing to achieve minimum bonum can be morally condemned by another for whatever effective measures (possibly involving summum malum elsewhere) it pursues to relieve that condition.
Life can be short and brutish and often is. I read (11.63) as the analogue of Hobbes's principal law of nature adjusted to the terms of this
primer. But of course that makes (11.63) a minimal constraint on morality, hardly a maximal one, and (against Hobbes) a conditional constraint, a "second-best" constraint, one that cannot be made out to be "critically universalized." That is, theorems (11.62) and (11.63) defeat the necessity of relying on first principles, or moral principles, or invariant reason ((2.2)). The argument from summum malum (or minimum bonum or similar sensibilities) is an argument that acknowledges that we cannot simply know, at the level of critical legitimation, what we ought morally to do. We act as reasonably as we can if, given the presumption of any critical morality, we construct ("parasitically") a morality committed to relieving summum malum or enhancing minimum bonum (or both) within the span of our sittlich sensibilities. My sense is that moral realists would never agree to that.
The upshot of (11.61) and (11.63), read in the context of the failure of moral realisms of the externalist sort, is simply this:
(11.64) if moral norms can be critically legitimated, then moral revolutions can be as well.
For, of course, the perception of summum malum or minimum bonum or some analogue of these, not suitably relieved or enabled by the prevailing Sitten of a society, may be made to yield new norms and values that would require the subversion and reconstitution of the offending structures of that society. For example, it might require the redistribution of land, the equality of races or sexes or different peoples, the repudiation of entrenched taboos, the empowerment of the dispossessed. By moral revolution, I mean: a critical vision, grounded in a sittlich way, that significantly exceeds the critical visions that obtain (in this or that viable society), prepared in the short run to favor the enabling praxical conditions by which, eventually, that vision may prevail in the prima facie legitimative sense. Hence,
(11.65) there is no principled disjunction between the moral and any other form of practical life.
Since, under symbiosis, there is no principled disjunction between theoretical and practical reason, the whole of human life has a moral cast, meaning, by that, that responsible human agents cannot ignore the ubiquitous relevance of questions about the scope or legitimative adequacy of their sittlich practices, their ideologies, their would-be critical
norms—in terms of possible moralities yielded by attending to summum malum or minimum bonum or the like. Correspondingly, the world has an economic, a political, a religious, a medical, and an educational "cast" as well, and moral agents are alert to the possibility of transforming their own lives conformably ((10.68)). Which is to say: to be a "civilized witness" need hardly be to serve a quietist vocation.
In my opinion, the single most important conceptual resource of a society (or cluster of societies) faced with the perceived intractability of its (their) sittlich world, with regard to critical norms (for instance, summum malum or minimum bonum ), lies with redefining (for cause) its prudential and prima facie goods and reforming its sittlich perceptions accordingly. Of course, to be effective, such a change must already be incipiently sittlich . Thus, in our own time, many believe (I think correctly) that the widespread use of heroin and marijuana is effectively uncontrollable, in the United States, by the usual societal sanctions and that the social costs of futile measures to control their use are unacceptable; that the same is true of putatively deviant sexual practices; that our evolving technology subverts and threatens to make obsolete "traditional" views of abortion, assisted suicide, "heroic" medical measures, a normal span of life, population growth, ecological control, and the like. We cannot foresee what, in time, will be judged morally legitimatable (in the critical sense)—whatever our present convictions—apart from the changing sittlich condition of actual societies.
Accordingly,
(11.66) societies may, reasonably, collectively promulgate a policy of nullum malum .
By nullum malum, I mean a society's affirmation that a part of its sittlich world, hitherto critically condemned—judged morally unacceptable or indefensible or the like—is now relieved of that onus. The usual rationale would involve emphasizing the ineffectiveness of relevant sanctions in real-world terms, social costs and other adverse consequences for summum malum and minimum bonum, the incipiently sittlich standing of such affirmations, the prospects of the coherence of relevantly altered perceptions and practices. For instance, homosexual intimacy, same-sex unions, medically managed drug use, and assisted suicide may possibly be "affirmed" as being (now) in accord with nullum malum . By affirm or promulgate, I mean, to act deliberately to alter our lebensformlich practices so as to reduce, possibly to eliminate, the prima facie condem-
nation of relevant behavior and life-styles of the sorts mentioned. The rationale for doing so is the conviction that
(11.67) there is no invariant rule or criterion of nature or reason ((2.1)) by which to confirm that anything human beings do, or can do, is, intrinsically, contrary, in the moral sense, to nature or reason—
which, of course, is not to deny that within the space of what is thought to be "natural" or "rational" (artifacts of our symbiotized world), given practices and behavior may be critically condemned, if any practices or behavior can be.
It does mean, however, that the usual treatment of legitimative strategies in moral matters—that between cognitivist (or "realist," that is, Aristotelian) and noncognitivist (or "constructivist," that is, Kantian) options as exclusive and exhaustive of each other (as Rawls argues)—is utterly inadequate. Within the terms of cultural realism (but only there), moral legitimation is realist; within the terms of symbiosis (but only there), moral legitimation is constructivist. If they are, then only a "second-best" morality is possible: a policy of nullum malum is then unavoidable. But that is to say that morality, like human life in general, is "existential"—never adequately guided by the presumptions of a formal or invariant system—and, of course, it says that a relativistic logic cannot be avoided.
Let me add a further, final thought. There is ample evidence in Kant's first Critique (pointedly in the preface to the second edition) that, although a constructivist, Kant is drawn to (what, following Peirce, I call) Secondness. The theme persists among the post-Kantian Idealists, who, in addition to being constructivists, are also committed to some form of symbiosis—which Kant is not. It is in this sense that, in accord with theorem (4.11)—to the effect that "external" relations between subjects and objects are artifacts posited within a symbiotized space—that the physical world is "metaphysically abstracted" from within the space of an Intentional world. For, as I have argued, the entire universe is an Intentional space ((9.23)). But if that is granted, then, on the argument, there cannot be a principled difference (as far as cognitive competence is concerned) between the legitimation of an objective natural science and the legitimation of an objective morality. There can be no basis for legitimating science without providing for legitimating morality at the same time. That, I suggest, is an extraordinary gain, one that we clearly
owe to Hegel. I have had the benefit of that in mind from the start of this primer.
Nevertheless, the difference between the legitimation of a coherent science and a coherent morality remains in place. I have shown how considerations of Secondness assist us (even in an-archic terms) is legitimating scientific objectivity, relative to the usual interests of prediction, explanation, technological control, and the like. I have been arguing in a parallel way with regard to moral objectivity: featuring, in the moral world, Sittlichkeit rather than Secondness and prima facie and prudential interests rather than prediction and technological control. I found it necessary to fall back to "second-best" considerations (and the consequences of doing so, regarding practical conflict, the logic of moral judgment, and ideological factors bearing on "the meaning of life"). And I have shown how policies of summum malum,minimum bonum, and nullum malum may be invoked to yield a "second-best" plausibility (objectivity) to competing, even conflicting, moral visions—which thereby escape the conceptual traps of moral realism and moral skepticism.
Now, the very need to fall back to a "second-best" morality helps to strengthen the ("parasitic") objectivity of certain candidate doctrines. For, given the inherent limitations of "critical" legitimation and its inseparability from "prima facie " (logically trivial) legitimation, it is impossible to disallow, on objective grounds, the following regulative policy:
(11.68) moralities should aim at increasing the scope of nullum malum within the space of their own Sitten, particularly when guided by summum malum, minimum bonum, and presumptions of what is required by, or is contrary to, "nature ."
Theorem (11.68) is, I suggest, the least encumbered advice (both morally and rationally) that follows from applying the notion of being a "civilized witness" to the entire foregoing argument—particularly if we concede (11.67). I don't deny that (11.68) may be opposed. But I cannot see how its validity can be dialectically disallowed, on objective grounds—that is, as a "second-best" morality competing with others. Any morality in accord with theorem (11.68) is what I should call a generous morality which, on non -moral grounds—given the sittlich encumbrances on a critical morality and the inescapability of invoking a "second-best" morality—cannot be bettered in terms of restricting the
ideologically arbitrary. (I should say that a "generous" morality was superior to what, following Rawls, has usually been understood to be a "liberal" morality. But that is another matter.) Its best application lies in disarming sittlich notions of what is said to be "unnatural" (morally) or "against" nature in the way of personal life-styles and what is said to be "ineluctably" required in the way of "natural" rights.
If, now, you review the argument of this last chapter, you will see that, in providing an account of objectivity and critical legitimation for morality under the conditions of symbiosis and historicity, I have in effect provided a paradigm for objectivity and legitimation regarding every and any sustained inquiry that we might undertake—those, for instance, of the sciences. For, every inquiry is, or is abstracted from, a human science ((9.58)), and every inquiry, like that of morality itself, is (as has just been argued) symbiotized and historicized. But to grant that is to grant:
(11.69) there is no principled distinction between theoretical reason and practical reason that accords with the canonical view (Aristotle's, for instance)
for: (i) reason is an artifact of history (ch. 8); (ii) there are no de re necessities ((2.6)); (iii) reference and predication, which are common to theoretical and practical discourse and inquiry, are historically entrenched ((3.8), (3.12)); and (iv) objectivity and legitimation with regard to truth and practical norms are, in a symbiotized world, of equal standing epistemically and metaphysically (ch. 11). Seen thus, the critical legitimation of scientific objectivity accords very well with the lebensformlich practices of science, relative to acknowledged salient interests (predictive power—for instance, technological success). The very same is now true of morality, except that eligible interests (summum malum, minimum bonum ) are more profoundly contested than those of science. This leads us in the direction of relativism. The objectivity of every other inquiry is, accordingly, open to legitimation along related lines—but not otherwise. To admit all that, however, is to confirm the amplitude and adequacy of the conceptual resources of the an-archic vision. That is:
(11.70) critical legitimation, applied alike to practical and theoretical norms, cannot exceed a "second-best" objectivity.
Under flux, under symbiosis, under historicity, it could not be otherwise. But I must break off here.