Issues and Reflections

Henry Armstrong was a partisan observer, but there is much truth in his advocacy of his mentor:

Kolbe is chiefly known to-day [in 1930, but the point is just as valid two generations later] by his attacks on Kekulé and van't Hoff: the fundamental importance of his work is overlooked. Some day when—as Sir Thomas Browne might urge us to do—we omit the "improperations and terms of scurrility" which he launched at Kekulé, he may come to be regarded as the parent of the modern system of resolved structural formulae: we here have to thank him for having made Frankland, whose senior he was, what be became.^[1]

No chemist was a more skillful or fruitful worker in the middle decades of the last century than Kolbe, who was in the very forefront among those investigating molecular "constitutions" and devising some of the earliest synthetic methods in organic chemistry. Moreover, it was precisely these directions—structure and synthesis—that proved to be the key to the great burst of scientific and technological innovation described in our story. Even though it was the true "structuralists" such as Kekulé, Butlerov, Crum Brown, Frankland, and Erlenmeyer who formulated and developed the theory in its essentially modern form, Kolbe's pioneering work needs to be recognized as such.

In pedagogy, too, Kolbe played a leading role. It is difficult to disentangle Kolbe's contributions from those of his predecessors, especially Liebig and Wöhler, but it is a truism that the German system of science education which all of these men helped to develop was gradually

exported (with admixture of important indigenous elements) to Britain, France, the United States, Russia, and other countries. This borrowing became quite self-conscious beginning in the 1860s, when German superiority began clearly to be sensed abroad; government commissions in Britain, Russia, and France set off alarms and produced reforms. Liebig was particularly influential in Britain, and Wöhler was the most prolific chemist breeder for the United States. However, characteristic Kolbean elements can be discerned especially in Russia, where Kolbe was the chemist of choice for komandirovky in the 1860s and 1870s,^[2] and in England, through the mediating influence of the Kolbeans Frankland and Armstrong. Armstrong was the leading English science educator at the end of the last and the beginning of this century and did much to spread enthusiasm for what he called the "heuristic" system. By this term, he meant that students should not be passive receptacles for knowledge by listening to lectures and memorizing facts (what he called the older "didactic" or "chalk and talk" method) but rather they should create their own knowledge by personal discovery and with their own hands, through a combination of intensive expert-guided and self-guided experiments. Much about Armstrong's teaching philosophy appears to come straight from the Kolbe laboratory in Leipzig.^[3]

Kolbe's mentors, colleagues, and students well appreciated the high significance of his work. The three heroic chemists of the "classical" generation—Liebig, Wöhler, and Bunsen—all agreed that Kolbe was the very best among the younger generation, regarding him above even Hofmann and Kekulé. (Liebig favored Hofmann during the 1840s and 1850s, but then their relationship cooled somewhat, after which Liebig transferred his primary loyalty to Kolbe.) Evidence for Wöhler's regard is his attempt in 1862 to persuade Kolbe to edit a new edition of his Grundriss der Chemie .^[4] From the beginning of their relationship until Kolbe's death, Bunsen's loyalty never wavered. Moreover, all three chemists strongly (though silently) sympathized with Kolbe's attacks on structural formulas and structural chemistry. In addition, Kolbe students such as Ernst von Meyer, Ost, Schmitt, Armstrong, Volhard, Graebe, Crum Brown, Curtius, and Beckmann prized their relationships with their mentor and knew well the important historical role that Kolbe had played.

In retrospect, however, the most salient aspect of Kolbe's career is the thoroughness of its destruction. As we saw in the last chapter, the beginning of the end was Kolbe's public exit in 1871 from the Deutsche

Chemische Gesellschaft and also, as Lothar Meyer put it, from the "Gesellschaft gebildeter Chemiker,"^[5] over a really minor matter that Kolbe managed to whip into a storm. Two years later, upon the death of Liebig, a committee comprised of the most prominent German chemists was set up to plan commemorative events. Kolbe was pointedly excluded, even though he was apparently at that time the leading candidate to succeed Liebig; this exclusion seems to have been due to the action of the chair, Hofmann.^[6] Upon Wöhler's seventy-fifth birthday (which was also his fiftieth anniversary of teaching), an elaborate appreciation was given him, but again Kolbe was snubbed, even though Kolbe was Wöhler's oldest student of any prominence and all were aware of Wöhler's regard for his former student.^[7] Even many of Kolbe's hitherto closest friends, such as Hofmann, Volhard, and Frankland, became alienated during the last years of Kolbe's life. After 1873, Liebig and Eduard Vieweg were no longer around, and those who still supported him, especially Wöhler and Bunsen, refused to do so publicly. In his last decade, Kolbe was isolated from the collegial community in an extraordinarily complete fashion.

Dramatic decline in the influence of a hitherto major figure in a field is a familiar story in the history of science, and such decline often takes place, as with Kolbe, in middle age. In many cases, this occurs through such obvious phenomena as psychological burn-out, physical exhaustion, or the complacency bred by a successful career—Kekulé, for example, suffered from all of these and published little personal research after the age of forty-four. What is unusual about Kolbe's story is the degree to which he remained engaged with both the theoretical and experimental sides of the science until the age of sixty, and the degree to which empirical evidence, including even many of his own experiments, failed to support his scientific beliefs. These factors transform an otherwise prosaic case into an interesting anomaly.

Some causal elements of Kolbe's tragic dénouement can be suggested from social and psychological factors. Kolbe's simple rural upbringing contrasted with the upper-bourgeois urban origins of such men as Hofmann, Kekulé, and Baeyer; in this sense he found a natural counterpart in Frankland, who had a similarly modest background. As Armstrong commented, he was plain spoken and straightforward, not at all eloquent or cosmopolitan. He must have bitterly resented Kekulé's dismissive and elitist responses in some of their polemics, for, whether conscious or not, that sort of treatment succeeded in accentuating the social distance between them. Kolbe's rejoinder was to highlight Kekulé's putative linguistic solecisms, a tactic that was intended to demonstrate a lack of Bildung . As time passed, Kolbe became ever more proud, imperious, and distant, and he lost the mental

and emotional elasticity to adjust to change. Moreover, his sincere religious faith, increasingly conservative politics, and open prejudices rankled with the generally centrist, materialistic, and agnostic peer community. In Germany there was nothing like the clear center-periphery dichotomy that was (and is) so prominent in French and British social geography, but Kolbe's career trajectory as a sort of satellite orbiting the dominant state of Prussia is remarkable, and we find in Kolbe a decided aversion to things Prussian and Berliner.

Psychologically, there appear to have been pathological elements, especially after 1870. Kolbe was nothing if not conservative in his theoretical preferences,^[8] and he began to view novel developments in chemistry as just another aspect of modernism. Somehow he began to associate structural formulas with sensualism and materialism, possibly even with irreligion. His whole life was devoted to the science of organic chemistry, and he saw that science almost in the personification of a pure virgin being seduced and destroyed by meretricious villains, by liberals, social democrats, traitors, atheists, Catholics, and Jews. In the 1850s and 1860s, he suffered periods of paranoia and severe depression, and after 1870 he appears to have had delusions of grandeur. He became a caustic critic of modernity tout court, a personality type that was by no means unique in Bismarckian society—as we saw in the preceding chapter.

What perspective on our subject is provided by simple considerations of generational change? It is remarkable that Kolbe's birth year (1818) distinguishes almost without exception nonstructuralists from structuralists in Germany: virtually all of the former were born before this date, the latter all after.^[9] This means that nearly all of our modernist protagonists were in their twenties and thirties during the crucial decade of the 1850s. It is also interesting to note the high degree of antipathy to theory in general expressed by the leading chemists of the older generation. Wöhler and Bunsen both were non- or even anti-theoretical their whole lives, and Liebig and Dumas ostentatiously rejected theory around 1840; Williamson and Wurtz also retreated from the theoretical realm. After about 1860, Liebig and Wöhler ceased to make even the most perfunctory attempts to remain current with the literature. They were viscerally repelled by the new system of formulas and had no understanding for them. In contrast, by the mid-1860s nearly all theoretically active (that is, younger) German chemists had become structuralists. In this sense, we can say that Kolbe was both typical and atypical of his generation: typical for being on the nonstruc-

turalist side of the cusp, even if barely, and thus finding most of his contemporaries and older colleagues agreeing with him in rejecting structural ideas, but atypical of the same peer group in being profoundly interested in theory, and virtually unique in being a non-structural organic theorist of any age in Germany in the years after 1865. To put the matter more simply, everyone in his older peer group other than Kolbe rejected structure theory simply by rejecting all organic-chemical theory.

A similar ambiguity is encountered when one examines the issue of style of theorization. Chapter 10 explored the suggestion that a self-conscious hypothetico-deductive theoretical style began to displace Baconian inductivism in Europe in the 1840s and 1850s and asserted that Kolbe was one of the leaders of this trend, at least in the chemical community. However, a concurrent trend was also in evidence, namely, an increasing tendency toward explicit use of conventionalist theory, and this Kolbe tenaciously resisted. Proceeding from a standpoint of naive realism, he objected that the structuralists' radicals were "idealized entities invented for convenience," whereas his radicals were "facts." The carbonic acid theory was indeed a kind of type theory, he averred, but his types were "real," while those of his opponents were merely "formal," "conventional," and "ideal."^[10] This predilection for manipulating conventional entities, he averred, had led modern chemists into making statements about matters that were simply beyond human ken.

The goal for which Kekulé strives, and which he considers accessible, is actually even more inaccessible for us than the moon, for we can see the moon and determine its form, but atoms we cannot see, and their form is perceivable with none of our senses.^[11]

The structuralists agreed that atomic arrangements are invisible to the physical eye, but responded that we must use our mental eye. "Such prophetic vision," Kolbe acidly commented, "has not been given to me."^[12]

On the rhetorical level, this sort of language is clear enough, but Kolbe got into problems of self-consistency in attempting to instantiate his points. One way in principle of distinguishing real from conventional entities might be to demonstrate genetic interrelationships for the former versus merely schematic interrelationships for the latter. Despite the emphasis in his fundamental paper of 1860 on genetic chemical relationships (usually substitution reactions) as the basis of his theory, many of his type assignments were not founded upon direct interconversions of one compound to another, and Kolbe rec-

ognized and conceded that fact.^[13] A related tactic might have been to operationalize his theoretical entities, demonstrating their more direct relationship to macroscopic experiments and concrete data. Kolbe certainly believed that he could make this sort of case, but in fact his radicals (oxatyl, carbonyl, hydrogen peroxide, methine, ethyl, and so on) were no more isolable than those of his opponents. Kolbe conceded this point, as well—ironically, in the course of an attempt to indict structuralism in a different sense. Structural formulas are mechanical and coarsely sensual, he wrote, a symptom of the modern "crassly materialistic treatment of scientific matters; the latter ought to be conceived by the mind and not mechanically."^[14] In sum, Kolbe tried but failed to make ontological and epistemological distinctions between his and his opponents' theories; what remained was nothing more than an objection to conventionalist rhetorical style .

Kolbe had little sympathy or understanding for physics or for the field of physical chemistry that was beginning to mature in his last years, partially because these disciplines also used conventional theory and sought to investigate the (apparently) empirically distant micro-world. It was appropriate that it was Lothar Meyer, the physical chemist and protégé of the early conventionalist and hypothetico-deductivist Franz Neumann, who remonstrated most effectively with Kolbe on exactly these points:

Why do you, you who have contributed so much to the correct understanding of the chemical constitution of compounds, wish to erect an a priori barrier for the research of yourself and others at a spot which we have not yet reached, and therefore cannot know whether a navigable channel or an impenetrable wall of ice stands before us? You say we will never attain knowledge concerning the spatial arrangement of atoms. I ask, why not? When the microscope was perfected we learned to see things of which we had no inkling, and also these days we measure things that are far too small to be seen even with the most powerful microscope. Why do you want to lame the wings of research with a categorical "To this point and no further!"? Did such a warning tablet succeed in sparing the so-called vital force from analytical dismemberment? . . . If only we are careful to beware of the (unfortunately quite frequent) confusion of our hypotheses and theories with the absolute truth, which we have perhaps never yet found, then speculations surely hurt very little, even the very boldest ones.^[15]

Frankland had similar words for his friend:

I cannot coincide in your condemnation of the use of constitutional formulae by young chemists. It seems to me that chemists who make no use of their imagination (Odling & Brodie for instance) do but very little orig-

inal work. The hard dry facts of the science are not likely to excite the enthusiasm of anyone. . . . The real progress of science depends upon the accession of new facts, and theories are only of use as incentives to discovery, and, looking back at chemistry during the past 25 years, it really does not seem to make much difference, as regards progress, whether the theories, which act the part of stimulants, be true or false.^[16]

Both Frankland and Meyer understood that they were speaking to a passionate theorist and that the real point in dispute was conventional-ism, whereas others, even to this day, have mistaken Kolbe's position by labeling it radically empiricist. He was, in fact, strongly inclined toward hypothetico-deductivism and philosophical realism; the former fit in well with the second half of the nineteenth century, but the latter did not.

Here we may have one means of understanding his (and many of his older compatriots') revulsion for structure theory. Benzene theory during the last third of the century can stand as a case in point. To a remarkable degree, Kekulé's cyclohexatriene structure was distrusted ontologically, while at the same time used in a confident heuristic fashion by most theoretically active organic chemists. The apparent contradiction is resolved by viewing most structuralists as (conscious or unconscious) conventionalists.^[17] More generally, structure theory did not make much sense from the standpoint of well-developed global theories such as electromagnetism or Newtonian mechanics, which may have increased the attraction of the conventionalist standpoint for structuralists. Most of them understood that structures proposed in the literature were usually partial and provisional, especially for complicated molecules. Such provisional structures were gradually refined and improved, often by their original proposer. This gradualist approach continued to characterize synthetic organic chemistry well into the twentieth century, indeed to a large degree even today. It contrasted, however, with the older style of proposing and then "proving" a particular point of view, and the new style stuck in the craw of Kolbe—as also of Liebig, Wöhler, and many other older chemists.

We can summarize all of this by affirming that issues of generational change offer us some helpful perspectives on the enigma of Hermann Kolbe. Significantly, however, Kolbe's actual patterns of theoretical manipulation—aside from questions of methodological style, rhetoric, and language—were essentially the same as those of his opponents. His theoretical entities were not more easily operationalized to the macroworld of the laboratory nor were they more thoroughly founded on genetic chemical relationships. For all of his emphasis on definitive "proofs," not one of his peers exhibited more twists and turns along

the theoretical pathways than he, nor more of an emphasis on hypothetico-deductivism, nor more heuristic use of chemical formulas, even if those formulas were different from those of the structuralists.

Kolbe also exhibited far more flexibility during the course of his early and mid-career than has often been supposed. He kept track of the literature and heeded evidentiary arguments, and during the 1840s and 1850s, he continually modified his positions in accordance with the advancing state of the art. These events have been detailed in chapters 3, 6, and 8. The position he had arrived at by 1860 was largely equivalent to that of the newer type theorists. Kolbe vehemently denied the arguments of both friends and rivals that he had joined rather than defeated his enemies, but the critics were essentially right. Using his "carbonic acid theory," Kolbe was able to make fundamental contributions to organic chemistry throughout the 1860s.

Nevertheless, the vestiges of electrochemical dualism retained by Kolbe resulted in some real differences between his and the structuralists' ideas: Kolbe avoided the idea of anisotropic bonds between atoms and affirmed strictly hierarchical molecular constitutions. These differences led to predictions and retrodictions that in certain cases were empirically distinguishable from those of the structure theorists. As we have seen in chapters 9, 12, and 13, in every such distinguishable case Kolbe failed to produce empirical confirmation of his own theory or refutation of the structuralists', and the cumulative anomalies in his position became ever more serious through the 1860s and 1870s. The remarkably one-sided character of these results helps to explain why structure theory became increasingly popular and powerful^[18] and why Kolbe succeeded in persuading not a single prominent scientist of the advantages of his variant, not even his own students. The real mystery of the matter is why Kolbe himself was not converted to structuralism and why he retained his theory of 1860 until the end of his life. Ultimately, the tentative social and psychological rationalizations suggested above must suffice.

The previous two sentences are undergirded by the assumption that evidence matters to scientists and, if sufficiently compelling, will prompt conversions even from a comfortable and well-loved theoretical position to one that is novel and initially distasteful. A generation ago such an assumption would have been regarded as unsurprising, but as historians and other people know, times change. A central and now well-accepted element of post-positivist philosophy is the claim that theories are in principle seriously underdetermined by empirical evidence.

This thesis has been expanded by advocates of the "strong program in the sociology of knowledge" to the position that scientific knowledge, like any and all other forms of human belief and doctrine, is entirely socially constructed, fully independent of empirical investigations.

In coining the quoted phrase and defining the program, David Bloor was careful to specify that causes other than social might in particular situations be determinative. This appears to allow entry to the influence of empirical evidence. In practice, however, the social constructivists tend to proceed as if no such influence were ever important. In his quest for Bloor's injunctions toward "symmetry" and "impartiality," Bruno Latour was moved to the position that "nothing extraordinary and nothing 'scientific"' ever happens in laboratories, and so we must "abolish the distinction between science and fiction." Similarly, Harry Collins believes that "the natural world has a small or nonexistent role in the construction of scientific knowledge." As Barry Barnes and Steven Shapin put it, "The intense concern of earlier generations with the special status of science and its allegedly distinctive characteristics has begun to ebb away."^[19]

The strong programmers have been met by strong criticism, including charges of purveying "voodoo epistemology" and the "pseudo-science of science."^[20] The social constructivists have also been known to use terms of opprobrium, referring to the "paralysis" of philosophical minds and the "bleatings of recusant epistemologists and methodologists."^[21] Difficulties with the positions of some sociologists of scientific knowledge include a tendency to overshoot in their correctives, a proclivity toward the construction of straw men, and their generally unconvincing responses to a variety of reflexivity problems. With regards to the first of these issues, no one, I think, would deny that the overall historiographic changes in the profession over the last generation have been highly salutary, providing a more contextual, more realistic, and more fully historicist vision of the past. Nearly all historians of science advocate the underdetermination thesis and regard social causation as a central element of historical explanation. But admitting underdetermination does not mean that the resulting gap must necessarily be filled exclusively by social causation, and even gross underdetermination does not mean zero determination.

Moreover, the assertion of a distinctive character to science need not be taken in the way that some positivist-Whig historians once viewed it, as a sort of purely cerebral and bloodless pursuit of onto-logical truth. Such a caricature is often the model attacked by strong programmers, but in fact modern historians and philosophers have long been constructing much more sophisticated and carefully nuanced conceptions of the nature of the scientific enterprise. The apparent

general conformability of the world with our socio-culturally based investigations of it does indeed suggest a certain epistemological asymmetry. I believe that there is in fact a degree of asymmetry here. A central paradox then requires resolution, namely, how those thoroughly socially grounded investigations can nonetheless exhibit certain special characteristics—success, power, predictive heuristics, technological applications, and ultimately what appears to be gradual ontological progress—that we have come to associate with science and that are dismissed by social constructivists. This problem is still an open and difficult one.

But the position of the constructivists also contains paradoxes. For instance, a thoroughly symmetrized approach conceals an implicit scientism more radical than any they are attacking, for many of the strong programmers confidently assert that all knowledge, including social relationships and social epistemology, is as transparent to human reason and empirical effort as the scientists' far more delimited territory. But ironically and paradoxically, in order to be consistent, they must at the same time relinquish all claims to have succeeded in erecting anything other than a socially determined castle of cards. Such circularities and reflexivities have often been discussed, both by advocates and by opponents of the sociology of knowledge movement.^[22] Many, including the present author, believe that the constructivists' attempts to show that such circularities are not vicious have not been successful.

In this book (especially in chap. 6), I have portrayed the "quiet revolution" that played out in Germany in the 1850s as a major change in the science of chemistry that was consummated surprisingly quickly and easily, hence quietly. There was so little controversy, I suggest, mostly because the force of evidence drove conversions efficiently, often in the face of social and career interests. This was even true for Kolbe himself. No one had more contempt for the French or their theories in the late 1840s and early 1850s than Kolbe. However, the striking new reactions and brilliant arguments by Williamson, Gerhardt, Wurtz, and Frankland during the early 1850s that convinced most of Kolbe's German colleagues to accept the French-English theories were by no means lost on Kolbe either. We have seen that by 1857 he had developed a theory of his own that was strikingly similar to the Williamson-Gerhardt newer type theory. Kolbe's pathological prejudices had failed to prevent him from understanding and being persuaded by the hated French ideas; they had only operated to prevent him from believing he had adopted them. He never became a card-carrying structuralist, for which one needed to relinquish all vestiges of electrochemical dualism; but he was a de facto fellow traveler

for a great part of the journey. Later, his attempts to stir controversy and to win support for his variant position were quite unsuccessful.

In according a determinative role to the impact of new empirical evidence and reasoning as explaining conversions to this revolution and thus urging, by extension to the general case, a stronger role for purely cognitive factors in the process of scientific change than is granted by many social historians of science, I do not advocate a return to the bad old days of untrammeled internalism. Indeed, sociologists of science, including no doubt the social constructivists, could—and, I hope, will—have a great deal of fun with this case as well, despite its having been particularly suited for my thematic purposes. (If scientific theories are strongly underdetermined by empirical evidence, all the more so must be historical interpretations.)

One case of particular interest in this regard is France; by focusing in this study on the favorable reception of the Gerhardt-Laurent reforms in Germany, I have said little about why they made such slow gains at home. The differential reception in the two countries must have been due to social, cultural, and institutional factors, since the French had essentially equal access to the same papers of the early 1850s that so quickly converted their German neighbors. Indeed, as I have emphasized, they had even more direct access to the personal influence of the leaders of the reform, Laurent and Gerhardt, and subsequently Wurtz. Even Williamson, the key figure in the transformation of the evidence from circumstantial to decisive, was a Francophile. Why were the French so oblivious to this movement?

A full answer to this important question has not yet emerged.^[23] Clearly, the facts that Laurent and Gerhardt were social misfits (provincials, ardent republicans, materialists, and possibly both of Jewish extraction) and that they almost defiantly refused to play the conventional social games, had the result that their true merits were much slower to be recognized than might otherwise have been the case. There might even have been an element of psychological backlash, in which the Laurent-Gerhardt reforms were regarded, in the time of political reaction after their deaths, as "tainted" or even dangerous—for look what a tragic fate the protagonists shared! Moreover, after the reforms caught on in Germany—which, as I have argued, happened even before the death of Gerhardt—they could be viewed, perversely, as German, hence foreign and suspect. Nor can it be mere coincidence that the first French chemist to base his textbook on the work of Laurent and Gerhardt was also the republican and scientifically peripheral

Alfred Naquet. In addition, it appears that the positivistic tone of French science, noticeable since before the beginning of the nineteenth century and continuing for more than a hundred years, worked to the disadvantage of the obviously theory-based movement described here. Why French science had this particular complexion is, again, incompletely understood, but it must be due primarily to social and cultural forces.

Two final non-cognitive factors deserve mention. The decentralization that was so characteristic of Germany in the decades before 1871 has often been depicted as driving vigorous and ultimately beneficial rivalry between the various states. Healthy competition characterized German academic science and university administration, as we have seen at many points in our story and as has been noted by other authors.^[24] It was less prominent in France^[25] or in Britain. This kind of competition operated on many levels—local, regional, national, and international, as well as in matters of prestige, prosperity, institutional structures, military and technical superiority, and so on. German as well as foreign observers saw an important cause of Prussian dominance over Austria and France in the superiority of German science, propelled by a competitive research ethos. Whether this kind of case can be sustained or not, it would appear that the research ideal had declined in France during the decades that it had increased so powerfully in Germany, and after 1871 the French began aggressively to address the problem.

Finally, it behooves the author of a biography, at the risk of belaboring the obvious, to stress the contingency of history and to note that individuals can notably succeed or fail to exert powerful influence in their peer communities. For whatever reason, after the death of Gerhardt in 1856 France lacked a critical mass of reform-minded chemists who could make the case for the new ideas.^[26] Wurtz was professionally lonely, just as Kolbe was isolated on the other side of the debate and on the other side of the Rhine. More generally, the rise of the German research ideal so prized and emulated was associated with only a handful of leading actors.

I am far from alone in urging a flexible approach as regards the outworn dichotomy between cognitive and social history of science.^[27] Flexibility, pluralism, and an eclectic and empirical approach are important elements for success in science. I have argued in chapters 7 and 10 that these attributes were precisely the necessary ones to accomodate the rise of structure theory and that they were the ones that Kolbe

notably lacked after 1870. The same attributes are also beneficial for the historian. Social and cultural forces are powerful, pervasive, and efficacious, and so is the strength of ideas and evidence as pursued by conscientious scientists on the "agonistic field" in which one marshals resources. In the valuable perspectives provided by recent sociological studies, the power and vitality of scientific ideas and logic themselves, the constant regulating appeal to the empirical world, and the contingent influence of individual actions should not be underestimated.