Independence and Constraint: Good and Bad Grammarians and The Power of Convention

In Book 14, Chapter 5 of his Attic Nights , Aulus Gellius presents a curious spectacle. Two eminent Roman grammarians are engaged in a heated public argument, all but coming to blows over the correct vocative form of the adjective egregius : first one, then the other presses the claims of his rationalized account (ratio ) or rule (definitio ), then back again; no end in sight. With a shrug and a sniff Gellius withdraws: "But since . . . their competition was going on at quite some length, I did not judge it worth my while to listen longer to those same, well-known arguments, and left them yelling and battling." The incident is intended to amuse and appall, as each grammarian clings to his ratio as though his life depended on it, in a display of emotion and egotism at once unseemly and boring.

The episode, like so many in Gellius, involves a competition for prestige centering on the literary culture. Explicitly, the contestants are the two

grammarians, each with his professional authority invested in his own ratio . But Gellius's dismissal implies the larger competition played out in the Attic Nights as whole: Gellius and his learned friends versus the vulgus semidoctum , "the common run of half-educated men," to which the "half-educated grammarian" (semidoctus grammaticus ) belongs.^[73] In various vignettes the grammarians thrust themselves and their learning forward only to be embarrassed by their betters; they are consistently losers in the competition for a place in the "aristocracy of letters."^[74] This competition, its tensions, and its resolutions are perhaps most responsible for shaping the grammarian's relations as a professional teacher with men of culture at large and for defining the image of the grammarian in the literary tradition from the early empire onward.

In this section, then, we will briefly trace the origins and consequences of this competition. Along the way we will see how the early grammarians were able to claim a place in the world of liberal letters through their technical skills; how the social and cultural elite regarded the rise of the profession; how once the profession and its skills were established the elite coopted them; and how as a result the elite's conventional values limited the profession's independence.

Of course for competition there must be competitors. The grammarians were latecomers to the contest, owing their position to the gradual emergence of the Roman schools of liberal letters as institutions distinct from the family, where the education of the upper classes had long been embedded.^[75] Appearing at Rome from the late second century B.C. , the grammarians over the course of the next century slowly disengaged themselves from the great households to which they were formally tied as slaves and freedmen or on which as men of otherwise humble origin they were wholly dependent. Like the teachers of rhetoric, the grammarians began opening their own schools from about 100 B.C. and gradually became identified as the teachers of children in language

[73] Vulgus semidoctum: NA 1.7.17. Semidoctus grammaticus : 15.9.6.

[74] For the grammatici , apart from the passages cited above and below, see esp. NA 4.1, 5.4, 6.17, 8.10 praef., 13.31, 16.6, 18.6, 19.10, 20.10. For the most important exception, Gellius's teacher Sulpicius Apollinaris, see below, pp. 59-60. The place of letters in the larger competition for prestige and social standing among Gellius's contemporaries is well described by Champlin, Fronto 45ff., esp. 49.

[75] On the development of the grammarian's school at Rome, see recently Bonner, Education 37ff.; and esp. Booth, "Appearance." On competition within the aristocracy as a spur to the emergence of the liberal schools, see the sensible remarks of Wallace-Hadrill, Suetonius 30ff. For the idea of structural differentiation applied to education in a systematic account of institutional change, see Hopkins, Conquerors 76ff. The present discussion, less concerned with the reasons for the differentiation, takes as its starting point the fact that it did occur.

and literature.^[76] In general, we can say that structural differentiation had gone as far as it would go by the beginning of the first century A.D. By that time the grammarian's school was formally separate both from the family and from other educational institutions, the rhetorician's school within the field of liberal studies and the school of letters outside.^[77]

These changes did not occur without friction between old and new. In their earliest, most explicit form the tensions of competition had as their focus the new institution that impinged most on public life. In a decree of 92 B.C. the censors attempted to close the schools of Latin rhetoric on the grounds that they contradicted the "habit and custom of the elders" (consuetudo ac mos maiorum ):^[78] the mos , we can infer, was the novitiate of the forum (tirocinium fori ), the traditional form of apprenticeship for public life, through which the prospective man of affairs attached himself as a youth to an established figure, learning how to act and speak as he followed his model and watched him go about his business. By its nature the tirocinium was part of a closed and rigid system that monopolized entry to a civic career. Access depended heavily on the ascribed status of the participants and the connections of family and friendship, and its methods were informal, based upon the personal relationship between the younger and the older man. The schools of rhetoric, offering wider access to an important skill, threatened the monopoly and provoked the (fruitless) attempt at repression.

Probably because the grammarian's connection with public life was always less direct than the rhetorician's, similar tensions involving the grammarian's school were less dramatic in their appearance and longer in coming. In fact, the most overt and sustained reactions did not begin until the first century A.D. , when the grammarian had already settled into his institutional niche and had begun explicitly to stake out language

[76] On the freedmen and slaves as grammarians, see now Christes, Sklaven . For dependence and disengagement, see esp. the cases of Antonius Gnipho and Lenaeus (Suet. Gramm . 7, 15). On schools of rhetoric, see below, n. 78. On the date, and on the gradual identification with the education of younger children, see esp. Booth, "Appearance" 123f. Suetonius's statement (Gramm . 3.3) that at certain times (temporibus quibusdam ) there were more than twenty schools of grammar at Rome is unfortunately vague in its chronology, but it must refer to a period well into the first century B.C.

[77] On the different kinds of schools, see Chap. 1 n. 44; for limitations on the differentiation from the school of letters see "Some Variable Definitions" above.

[78] Suet. Gramm . 25.2, specifying Latin rhetoricians with their own schools (ludi ), thus different from the more general SC de philosophis et de rhetoribus of 161 B.C. (Suet. Gramm . 25.1; Gell. NA 15.11.1); on the opening of the schools of rhetoric in the first decade of s.I B.C. and for the interpretation of the edict of 92 B.C. that follows, see Schmidt, "Anfänge."

as his area of expertise: it is possible here to glimpse the causes and effects, as the profession's differentiation fostered and was reinforced by the development of a specialized skill, with repercussions in the literary culture more broadly. A grammarian's claim to stand against Tiberius and control Latinity by limiting words' "citizenship"^[79] and, in Nero's reign, Seneca's sarcastic reference to the grammarians as the "guardians of Latin speech"^[80] mark out the period when the grammarian became identified in his own mind and in others' eyes as the agent of linguistic control. It is not accidental that the first comprehensive ars appears in Latin during this same period, composed by the professional grammarian Remmius Palaemon and providing the model for future handbooks.^[81] Intellectual history here catches up with institutional and social history: the ars defined and codified the professional's expertise, the systematic analysis and the rules that were his special excellence, and so helped install the grammarian in a cultural system to which he was a newcomer.

In this conservative milieu, the grammarian's rules can still seem novel to Gellius a century later, when he turns an unfriendly gaze on "those who pay homage to the new-fangled conventions [nova instituta ] of the grammatici as though they were the sacred objects of sacred precincts []."^[82] Yet Gellius's impatient disdain here dearly has a meaning different from what he intended. His characterization really shows how far the grammarians had progressed: it recognizes both the validity the common run of educated men attributed to the new-fangled conventions and the prestige those conventions' makers had come to enjoy.

This common grant of validity and prestige had several results. If two grammarians battling over the fine points of morphology are burlesque figures, the comedy nonetheless has a truth at its core: the rival grammarians cling as if for life each to his own rule because their lives—insofar as these were identified with professional and cultural status—did indeed depend on the rules. At the same time, for literary men who would stand apart from the common run, the grammatici became figures with whom they must reckon and from whom they might distinguish them-

[79] See Chap. 1 n. 16; cf. n. 113 below.

[80] Ep . 95.65. See also Ep . 108.30ff., Seneca's contempt for the grammarians, as figures concerned only with linguistic detail; and cf. Ep . 58.1ff., on "wasting one's time with a grammarian."

[81] On the influence of Palaemon's achievement, see Barwick, Remmius 146ff., 236ff. On previous artes at Rome as early as the Sullan age, and so within a generation of the introduction of the grammarian's school, see Barwick, ibid. 109ff., 229ff., with, e.g., Bonner, Education 55.

[82] NA 17.2.15. Note also Gellius's contempt for turba grammaticorum novicia at NA 11.1.5; and cf. NA 16.7.13, isti novicii semidocti .

selves. Writing on the language as a learned amateur in the time of Nero, the elder Pliny could foresee that his work would provoke the grammarians, and he registered a combative glee when his prediction proved correct.^[83] A generation later, the literary guru Yalerius Probus self-consciously set himself apart from the ordinary professional grammarian: Probus had followers (rather than pupils), three or four of whom he would admit to his home of an afternoon (not meet in larger groups, in a classroom, in the morning), where he would recline (not sit in a teacher's cathedra ) and hold conversations (not deliver lectures).^[84] It is the picture of an intimate and elite coterie, gentlemen meeting in an aristocracy of letters. It is of a piece with this picture that Probus despised "those rotten rules and cesspools of grammar."^[85]

Several factors aggravated the tensions that developed over the course of the first century and continued into the next. First, there was the propaedeutic paradox: grammar had become and was to remain the first stage in a liberal education, the fundamentum of eloquence, in the metaphor repeated from Quintilian through Cassiodorus;^[86] as such, it was both niggling and necessary. At the end of the first century Quintilian still had to defend grammar against the view that it was "insubstantial and jejune."^[87] The reasons for this view are understandable. Seemingly removed from the concerns worthy of gentlemen by its immersion in minutiae, grammar required the mastery of those rotten rules; narrow and achingly technical, it summoned up the old distinction between the technician's specialized training and the broad culture of the aristocratic ideal and fell decisively on the side of the former.^[88] But because of the institutional norms and cultural expectations that had developed among the elite, grammar was also unavoidable. The grammarian, as guardian, controlled access to the language and the education one's peers valued.

[83] NH praef. 28.

[84] Suet. Gramm . 24.2, with Grisart, "Valerius" 385. On the connotations of "follower" (sectator ) in particular, see below, with n. 107.

[85] The phrases appear in the anecdote preserved at Gell. NA 13.2.1: finitiones illas praerancidas et fetutinas grammaticas .

[86] Quintil. Inst . 1.4.5; Cassiod. Var . 9.21.3.

[87] Quintil. Inst . 1.4.5ff.

[88] On the distinction, see the recent survey of Christes, Bildung 15ff., 196ff. The narrowness is part of, e.g., Gellius's contempt for the semidocti : for the narrowness of the grammarian vs. the broader culture of the ideal, see esp. NA 20.10. Note that Quintilian's review of the grammarian's tasks, Inst . 1.4-8, dwells on his linguistic instruction; the culturally more inclusive task of enarratio is considered only in the latter part of 1.8. Contrast Florus's attempt to put a fair face on the grammarian's professio litterarum , concentrating wholly on the ethical content of the instruction and, at least as his text is now preserved, ignoring the linguistic-technical: cf. Chap. 1 n. 5.

His institutional place by itself made the grammarian a consequential figure endowed with power and respectability as a cultural authority, and his mastery of the language gave him a conceptual power that even those who might despise his pedantry were unwilling to abandon themselves.^[89]

Yet the grammarians were not obviously or unambiguously gentlemen, worthy of this power: they were by and large stigmatized for taking fees,^[90] and—especially at Rome, in the early first century—they were still comparable in their humble social origins to the grammarians of the preceding century. As competitors, they were at once worthy in their cultural standing and unworthy in their persons. Remmius Palaemon is emblematic of such circumstances, a grotesque in the eyes of the upper classes. Born a slave, he originally learned the lowly craft of weaving (the story went), and got his first taste of letters by accompanying his mistress's son to school; he then taught as a freedman at Rome, where his school brought him yearly earnings equivalent to a knight's census, and an estate. All his wealth went to satisfy his taste for luxury and his lusts; because of his vices, Tiberius and Claudius warned students off his school, but without effect. Vergil, he claimed, had used the name "Palaemon" in his work (Ecl . 3.50ff.) because he had foreseen that Palaemon himself would one day be the greatest scholar of poetry. Varro—a Roman senator and the greatest scholar of Latin before him—he called a pig.^[91]

The sketch is familiar: it is the standard picture of the arrogant and depraved parvenu. With some adjustment, this image of Palaemon could be superimposed on the caricatures of wealthy and powerful freedmen from the first century, especially in the imperial service, or of the suddenly risen notarii from the fourth. In each case the picture appears when competition for honor has been aggravated by the inconsistent status of some of the competitors, men who might enjoy prestige in one

[90] Cf. Dahlmann, Kleine Schriften 256 (and add Suet. Gramm . 3; Dio Chrys. Or . 7.114); Hopkins, Conquerors 124; Booth, "Image" 5; and below, n. 103. In his comment on Suet. Gramm . 24.2, numquam enim ita docuit [sc. Probus ] ut magistri personam sustineret , Grisart was probably correct to conclude ("Valerius" 385 n. 26) that "la phrase veut dire que Probus ne fut jamais un professionel exerçant son art moyennant salaire " (Grisart's emphasis). On change and continuity in attitudes toward payment, see Chap. 3 pp. 122-23.

[91] Suet. Gramm . 23; cf. most recently Christes, Sklaven 98ff.

or more areas (letters and wealth for Palaemon, political office and power for the notarii ) but have none in others (birth in Palaemon's case, letters and birth in the case of the notarii ).^[92]

Which features of our picture of Palaemon might be caricature it is not my purpose to decide here,^[93] nor is it necessary to question individual details. The overall emphasis speaks for itself: it betrays the tendency of a tradition to move in the direction of cliché and to resolve awkward inconsistencies by imposing a single image. An illustration can be found close to hand. One can gather references to teachers of letters in the high literary tradition from Demosthenes to Libanius and emerge with the consistent picture of a thoroughly humble figure with no claims to status in any form, a "gagne-petit universellement méprisé."^[94] Nothing learned from these sources will prepare one to find, for example, a teacher of letters offering a substantial dedication to the goddess Leto in the name of the emperor, on equal footing with a rhetorician.^[95] This does not mean that the tradition is wholly wrong, still less that the literary references are simply a mindless commonplace. On the contrary, the example points to an important need the commonplace meets, offering the reassuring image of a world where the people one despises are really no better than they ought to be. In Palaemon's case, that reassuring image is preserved, at very least, by the proportions in the tradition Suetonius retails and by the shape it takes in the telling. In the long entry on Palaemon, only three terse clauses are devoted to his cultural achievements;^[96] all else is given over to his regrettable mores . His main claim to achieved status is thus submerged; his other potential claim, his

[92] On inconsistencies in status and their consequences for the imperial freedmen, see Weaver, "Social Mobility." On the notarii , see above "Some Variable Definitions" ad fin .

[93] Note that Suetonius's phrasing at the start of the entry (Gramm . 23.1: Q. Remmius Palaemon Vicetinus mulieris verna primo, ut ferunt, textrinum, deinde . . . litteras didicit ; cf. ferunt at ibid. 3 [below, n. 98]) should suggest that the account bears the stamp of gossip, and should therefore recommend greater caution than is usually apparent in its handling.

[94] Marrou, Histoire 484. For references to teachers of letters, see the collections of Marrou, ibid. 223; Booth, "Image" 2, "Some Suspect Schoolmasters"; and cf. Part II no. 241.

[96] Suet. Gramm . 23.1f.: sed capiebat homines cum memoria rerum tum facilitate sermonis: nec non etiam poemata faciebat ex tempore. scripsit vero variis nec vulgaribus metris . Note that even the phrasing here seems not entirely neutral: capiebat = "charmed," "won over," almost "seduced." With the brevity here, contrast the sketch of Valerius Probus, Gramm . 24, a much larger proportion of which is devoted to the man's scholarly activity.

wealth, is tainted by association with his inherent vices.^[97] In sum, he emerges as a man without weight or substance, ultimately laughable.^[98]

We could never gauge Palaemon's contribution to grammar from the tradition that Seutonius passes along to us; the main lines of that tradition aim to put the upstart in his place. He is a bad grammarian, and not in the sense that he is a bad scholar (as we would use the phrase, to mean that he is bad at what he does) but because he is a bad person—morally defective, unworthy of the trust placed in him (recall the warnings of Tiberius and Claudius) and surely unworthy of his spectacular success. Not for the last time, as we shall see, personal and ethical judgments play a significant part in determining how one is to regard the grammarian; insofar as the tale of Palaemon is remarkable, it is so because ethical judgment seems to play the only part. The matter is thereby simplified considerably.

Things would not, however, always be so simple, as the Attic Nights shows. The competition informing so much of that work is, to be sure, predictably resolved by the revelation that one's competitors—including many bad grammarians—are laughable and unworthy. But for Gellius, writing in the second half of the second century, the competition is complex: there are no blatant inconsistencies of status; the villains cannot easily be caricatured and dismissed as arriviste grotesques, and their claims to learning receive dose attention. Thus the portraits of the bad grammarians also become complex as their lapses in learning are drawn in to complement and highlight their lapses in mores .

This complexity has much to do with how the social composition of the profession changed, and that change in turn is a measure of how prestigious the grammarian's role had become. From the late first and early second century onward, the profession began to attract members of the respectable classes. This was a familiar process: the elite habitually claimed positions that were worth holding, "pulling the ladder up after themselves."^[99] The grammatici who appear in Gellius's work as despised rivals for a place in the aristocracy of letters would probably not have had social origins radically different from Gellius's own.^[100] The competition takes place more nearly among peers, which probably made it only

[97] His wealth is mentioned only to provide a measure for his luxuria: Gramm . 23.2, luxuriae ita indulsit ut saepius in die lavaret, nec sufficeret sumptibus, quamquam ex schola quadringena annua caperet, ac non multo minus ex re familiari .

[98] The entry ends (Gramm . 23.3) with a joke handed down (ferunt ) at Palaemon's expense, concerning his libidines .

[99] Cf. Hopkins, "Elite Mobility."

[100] On Gellius's vita , to be set in the middle range of good society, see Holford-Strevens, "Towards a Chronology." It is significant that although the grammatici are spared little in the NA , none is skewered for his social origins. (I assume here, as elsewhere in my comments on Gellius, that the details in his vignettes can claim verisimilitude when they are not strictly historical, and that when historical they are subject to the improvement of Gellius's literary art.) On the grammarians' origins in this period, see now also Wallace-Hadrill, Suetonius 38f.

more tense and difficult and no doubt influenced the strategy Gellius chose.

In anecdote after anecdote Gellius shows that whatever their claims or pretensions, most grammarians were nonetheless neither scholars nor gentlemen. Their main claims to cultural standing—their control of the language, and especially their rules—are repeatedly deflated, and their intellectual failure is usually combined with ethical lapse and social catastrophe. The grammarian who misguidedly trusts in his skill and arrogantly claims center stage is reduced to "blushing and sweating" before his betters.^[101] Or, as in the anecdote earlier described,^[102] the grammarians make an absurd and regrettable spectacle of themselves, not just engaging in an undignified public wrangle (that would be bad enough) but wrangling over something Gellius can dismiss as old hat. Or, in a direct confrontation with Gellius, an unbecomingly boastful fellow (homo inepte gloriosus ) must finally cover his embarrassed ignorance with a self-important (but self-destructive) defense, "That's no small question you ask; I don't teach that sort of thing for free": his desperate retreat into professional status is simply the finishing touch in a picture of cultural and ethical inferiority.^[103] Or, meeting another contentious grammarian, Gellius amuses himself by coolly making up his own rule (finitio ficta ) on the spur of the moment to suit his argument and sends the grammarian off with a flea in his ear: the rule is false, Gellius says, but you cannot prove it is.^[104] The implications are of course disastrous: so far as the grammarian's precious rules are concerned, anyone can play the game, with no authentic claim to validity; and without the claim, there remain only those rotten rules and cesspools of grammar that Probus had denounced, and the grammarian's silly and arrogant pedantry, which Greek epigrammatists had satirized at Rome still earlier in the first century.^[105]

Yet such vignettes tell only part of the story. They report easy victories over faceless grammarians, but they do not convey the importance that Gellius attributed to grammar itself. That was a serious business for

[101] NA 19.10.

[102] NA 14.5.

[103] NA 13.3.3; cf. 18.14.1 (the contempt for a iactator quispiam et venditator Sallustianae lectionis , who is contrasted with Sulpicius Apollinaris) and n. 90 above.

[104] NA 15.9.6ff.

[105] Cf. esp. Anth. Gr . 11.321 (Philippus), with 11.347; 11.138, 140 (Lucilius); cf. also 11.279 (Lucilius).

Gellius, as he demonstrates not only in his insistent attacks on the grammarians—if they were unimportant, he would not attack them—but also in the high proportion of chapters that are devoted to grammatical questions and show Gellius's readiness to use the grammarians' techniques and categories when it serves his purpose. How could he not, when he had been immersed in them through his schooling? The substance of grammar was just too important, in fact, to be left to the common run of grammarians, the bad grammarians. Accordingly, heroes must be summoned up to show how it should be done and to put the villains to rout; and the hero himself can be a grammarian, provided he is the right sort—a good grammarian, like Gellius's teacher, Sulpicius Apollinaris.

Sulpicius is of course shown to be more broadly and more deeply learned than the grammarian of the vulgus . More noteworthy, the difference is presented as much in terms of personal attachment and mores as in doctrina . Gellius first associates with Sulpicius not as a puer of the age normal for a grammarian's student but only after assuming the toga virilis , when he has gone on his own in search of "more expert teachers":^[106] the relations and status of student and teacher are immediately characterized as extraordinary thereby. Gellius is not a common pupil but a "follower";^[107] Sulpicius is "our friend Apollinaris," a familiaris ,^[108] contrasted at one point with a vulgar "peddler of Sallust."^[109] The bond is different from and more intimate than the exchange of cash for learning, the normal, tainted relationship between student and teacher. Further, the experiences of Gellius and Sulpicius are not set in the classroom but belong to the palace, the booksellers' quarter, or the bibliotheca Tiberiana^[110] and are thus part of the public or semipublic intellectual life of the city. Their experiences in fact suggest nothing so much as a metamorphosis of the old tirocinium fori . Nor is the relationship confined to Gellius's adolescence, since Sulpicius is there to give advice to the more mature Gellius,^[111] as he does to men still more eminent.^[112]

[106] NA 18.4.1, which of course implies that even at this still-tender age Gellius was able to distinguish the good coin from the base.

[107] Sectari: NA 7.6.12, 13.18.3, 20.6.1; for sectator elsewhere in the NA , see, e.g., 3.1.5, sectatores of Favorinus. Cf. above, with n. 84, on Valerius Probus. For the connotations of personal loyalty inherent in sectator , appropriate to members of succeeding generations within a family, see esp. Chap. 1 n. 74 (on CIL 6.1416, 1418); and cf. Anon. Pan. Lat . 7.23.2 Galletier, multi sectatores mei equated in the preceding sentence with the rhetorician's natural children.

[108] NA 11.15.8, 13.20.1.

[109] Cf. n. 103 above.

[110] NA 19.13,18.4,13.20.

[111] NA 12.13, when Gellius was a iudex .

[112] NA 13.18, advising Erucius Clarus; cf. 19.13.

Personally, where other grammarians are assertive and combative, Sulpicius is content not to press his authority:^[113] mild in rebuke and quietly confident when he knows he is right, he handles pretenders to learning with "the kind of wily irony that Socrates used against the sophists."^[114] "A man with a surpassing knowledge of literature," "a man adorned with choice knowledge," "the most learned man I recall,"^[115] Sulpicius could be taken for one among the learned amateurs, did not Gellius tell us that he was a grammaticus and magister .^[116] That, of course, is precisely the point: Sulpicius is the grammarian as gentleman. Consistent with this, his writings were belletristic epistulae ,^[117] not an ars or other technical tract.

In his portrait of Sulpicius and in his concern with things grammatical, Gellius shows us a stylized, idealized gentrification of the discipline: gentlemen and the grammarian as gentleman together immersed in grammar. If grammar had become established as part of Rome's literary world, much of Gellius can be read as an attempt to show how it should behave accordingly, with manners and learning alike refined and exquisite. The attempt is not unique to Gellius but recurs in the period with which we are centrally concerned.

Two and a half centuries after the Attic Nights , another good grammarian is sketched in the pages of one of Gellius's literary descendants—Servius in Macrobius's Saturnalia .^[118] As in the case of Sulpicius, Servius's excellence is expressed in terms of both doctrina and mores and is revealed in informal, personal relationships: he is "at once admirable in his learning and attractive in his modesty" (iuxta doctrina mirabilis et amabilis verecundia ),^[119] and both his learning and his verecundia are displayed in the symposium held by "nobles and other learned men" in their discussions of Vergil. The most striking feature here is that the virtue attributed to (or imposed on) the grammarian sets the tone of the entire work, which is itself the most profoundly grammatical product of the amateur literary tradition in late antiquity.

One of the cardinal virtues, verecundia can be translated as "modesty"; more accurately (if more cumbersomely), it names the sense of propriety deriving from a regard for the opinion of other men and an awareness

[113] Note esp. NA 19.13.3: Sulpicius graciously allows Fronto the authority to give citizenship to words—an evident alteration of the metaphor's original application (cf. Chap. 1 n. 16) that conforms to the ethos of Sulpicius.

[114] Mildness: NA 13.20.5. Irony: 18.4.1.

[115] NA 4.17.11, 16.5.5, 13.18.2.

[116] NA 7.6.12, 18.4.1.

[117] NA 15.5.3; cf. 13.18.2f.

[118] For the points raised in the following paragraphs, see further Kaster, "Macrobius."

[119] Sat 1.2.15.

of one's own position (especially one's hierarchical position) relative to others in a given context. It is the quality found, for example, in the deference an inferior owes to a superior (women to men, a young man to an older man, a humble man to an aristocrat), in the sense of shame that restrains a superior from humbling himself before an inferior, or in the awareness of parity that, ideally, checks competition between equals.^[120]Verecundia is the virtue of knowing one's place, the virtue par excellence of the status quo , an abundantly social virtue, regulating the behavior of men in groups.

In the ideal world envisioned by Macrobius, verecundia is so spontaneous as to seem innate. Thus Servius, who possesses a naturalis verecundia ,^[121] is found now deferring as a young man and a grammarian to his elders and betters, now offering a contribution as an expert, according to the propriety of the situation. So too the other guests as a group spontaneously take their places in a hierarchical rank (ordo )^[122] and know individually when to yield to others' expertise, when to assert their own, how to combine becomingly the two kinds of behavior. And Vergil himself is presumed to have exhibited precisely the grammarian's qualities in his own sphere, delicately coordinating deference and self-assertion in his treatment of the literary tradition.^[123]

This deep sense of propriety gives the Saturnalia its core and makes erudition a moral quality. The innate regard for others' opinions, the capacity for gauging how one ought to behave in general, and how in particular one should respect the cultural heritage others value provide the impulse to learning. That impulse is brought to fruition by another virtue, diligentia , the scrupulousness that in social relations characterizes the dutiful behavior of friends and in intellectual life maintains and deepens one's contact with one's culture and makes one truly learned.^[124] The model of excellence is the scrupulous reader (diligens lector ), exerting himself out of respect for the text, doing his duty by reciprocating the

[120] See Kaster, "Macrobius" 224ff.; with, e.g., Ambros. De off. min . 1.17.65ff., on verecundia among the officia adulescentis ; and esp. Jer. Ep . 66.6, where verecundia denotes the sense of shame produced by consciousness of rank, and the consequent difficulty for a member of the upper classes to descend to the life of the vulgus . Cf. Reg. orient . 17 (PL 103.479D): among the rules governing a praepositus of a monastery, the injunction ne perdat animam suam propter verecundiam . Cf. also Ambros. Apol. David altera 11.56: non mediocre autem quod Nathan denuntiavit ei [sc. David ], hoc est inferior propheta. gravis enim verecundia pudorque delictum ab inferiore reprehendi .

[121] Sat . 7.11.1.

[122] Sat . 2.2.1ff., 7.4.1ff., with Kaster, "Macrobius" 227ff.

[123] Sat . 1.16.44, poeta doctrina ac verecundia nobilis , with Kaster, "Macrobius" 231ff.

poet's diligentia ; conversely, the ignorant man (most notably, in the Saturnalia , the cross-grained aristocrat Evangelus) is such because he lacks diligentia , and he fails to be scrupulous because he lacks verecundia .

Learning thus follows mores : the learned man must first be virtuous; the ignorant man is necessarily depraved. As a result, the grammarians who are criticized in the Saturnalia , as in the Attic Nights , are vulnerable for more than their failures in learning. The "vulgar troop of grammatici " from whom Servius is distinguished, grammarians concerned only with their linguistic expertise, who neglect the studia potiora —religion, philosophy, antiquities, and other Realien —that Vergil incorporated in his work, are fundamentally moral failures.^[125] But a curious twist is involved here as well, one of the features that distinguish the Saturnalia from the Attic Nights , and Macrobius's Servius from Gellius's Sulpicius. The bad grammarians are among the villains of the piece, but Servius, the good grammarian, is himself apportioned only linguistic matters in the communal discussion of Vergil; the studia potiora belong to the noble guests. The grammarian's ethos and expertise stand at the center of the work, but the grammarian himself is at the margins: so thoroughly has grammar been engrossed by the learned amateurs of Macrobius's ideal that the group must even be reminded to include Servius in distributing their roles.^[126]

Later on we shall consider the behavior of real grammarians and trace the lines they followed in life between the idealized images just described: the good grammarian of Gellius, moving freely among men of power and standing, and the good grammarian of Macrobius, following in the noblemen's wake, "his eyes upon the ground and looking as though he were trying to hide."^[127] But in moving from the time of Gellius to that of Macrobius, one notices a fact that requires preliminary discussion here, since it concerns the development of the grammarian's profession. The grammarians have no lack of competitors and critics in the literary world of the early empire; in the later period the critics' silence is startling by contrast.

The sniping at grammarians in the Saturnalia is far less elaborate than in the Attic Nights (perhaps partially as a function of genre, and certainly because of Macrobius's greater concern to present a front of unity, compared with the open combativeness of Gellius).^[128] Beyond the passing hits in the Saturnalia and a similar swipe in Tiberius Claudius Donatus's Interpretationes Vergilianae (1.1.5ff. Georgii), criticisms or lampoons fall into three categories: a few scattered and largely sectarian or ad hominem

[125] See Sat . 5.19-22, 1.24.12f., with Kaster, "Macrobius" 235f., 252ff.

[126] Sat 1.24.14-20.

[127] Sat . 1.2.15.

[128] Cf. Kaster, "Macrobius" 247f., 259f.

sneers from one or another of the learned professions,^[129] a few limp revivals in Latin of the earlier epigrammatic tradition,^[130] and the remarkable poems of Palladas in Greek.^[131] A grammarian in his own right, Palladas yet turns his general contempt for the world upon himself and his profession, and falls upon the sword of his own satirical epigrams. And that is all—save for a few voices among a new group of potential competitors, the Christian Fathers.^[132]

Since from no point of view can the grammarians be said to have become unimportant to the literary culture and therefore too insignificant to be attacked, an explanation for this late-antique calm must be sought elsewhere. It is possible, for example, to point to cultural inertia: the grammarians' conventions were no longer new-fangled but had long since become familiar fixtures in the life of the liberally educated elite and could be valued as such, or at least be taken for granted. It is also possible that some tensions were dissipated by changes in the grammarians' social status as members of the respectable classes took over the profession. The trend visible at the end of the first century is certainly dominant by the fourth; although a few grammarians still emerge from humble levels of society,^[133] the other grammarians whose origins and circumstances are observable cluster around the curial order and belong to the landed classes. They are not categorically their students' social and economic inferiors.

Yet these observations are not sufficient to explain the absence of comment and criticism. Cultural inertia evidently involves a petitio principii , since it does not address the questions, Why did those conventions settle into such easy familiarity? Why, in fact, did so little change in the substance and especially in the conceptual bases of grammar over the centuries? The familiar conventions of course continued to serve the advantage of the grammarians, who therefore had little incentive to

[130] Cf. Epigr. Bob . 46, 47, 61, 64.

[131] See Part II no. 113.

[132] See "Polished Speech, the Common Good, and Christianity" below.

[133] See Chap. 1 n. 59 for the grammarians of libertine birth at Bordeaux. Note that the inconcinnity is still felt by Ausonius, although not as grounds for contempt: thus Prof . 21.25-28, on Crispus and Urbicus, ambo loqui faciles, ambo omnia carmina docti, / callentes mython plasmata et historiam, / liberti ambo genus, sed quos meruisse deceret / nancisci ut cluerent patribus ingenuis . On the grammarians' social status generally, see Chap. 3.

change.^[134] Part of the answer no doubt lies there. But that can still tell no more than half the story, since it overlooks the grammarians' acquiescent audience—the members of the educated classes. Nor can change in the grammarians' status offer a solution, since, as Gellius shows, competition need be no less intense when it takes place among men who are more nearly peers.

It is possible, however, to suggest a broader hypothesis that might provide, if not a solution, at least a framework for fruitful discussion. The grammarians' conventions stayed as they were, and the grammarians ceased to be disturbing competitors, because the profession's social circumstances limited its differentiation and autonomy. Despite their emergence as separate institutions, the schools—and perhaps especially the grammarian's school—remained partially domesticated: the profession's horizontal differentiation Was limited because the grammarians remained tied directly to the family, to the representatives of the gentlemanly amateur tradition, and to the values of both.

Although it is commonly said, for example, that second- or fourth-century pedantry represents a decadent acquisition of knowledge for its own sake, one could well suggest a different view of the matter: knowledge was pursued not for its own sake but as a predominantly social phenomenon, as an appanage of personal relations and a token of accepted virtues. Knowledge could not usually be pursued, analyzed, pondered for its own sake. Or to put it another way, the habit of speaking (anachronistically) of ancient universities ought to be avoided, not simply because it obscures substantive curricular differences but especially because it obscures the institutional differences and their consequences. Antiquity lacked the institutional buffer that is raised between the lay and professional worlds by the modern university, which serves as the seedbed of the learned professions; and as a result antiquity had no place where a profession could attempt to set its own course and determine its own values.^[135]

In the grammarian's world autonomy and dependence, achieved and ascribed status, professional skill and the virtuous regard for the opinions of (nonprofessional) others rubbed shoulders. The distinction—so basic to the modern notion of a profession—between impersonal evaluation

[134] See Chaps. 4 and 5, esp. Chap. 5 pp. 196-97.

[135] This does not of course imply that restraints on the institutional and personal autonomy of the professions cannot be found in the recent past or today, or that the modern professions are context-free; rather, the embedding takes different forms in different contexts. For studies of the problem in specific settings, see Clark, Prophets ; Keylor, Academy ; and esp. Ringer, Decline , and Weisz, Emergence . Much light is also thrown on the question by Bledstein, Culture . For a recent general discussion, see Ben-David, "Organization."

of skill and personal favor or antipathy had no central place in the dominant ideology; instead, the distinction was blurred by an ideal that subordinated skill to ethical qualities. Old patterns of behavior persisted long after the profession's formal establishment: the face-to-face relations that characterized the dealings of grammarians with their patrons in the great households of late republican Rome never fundamentally changed, regardless of the different ways those relations came to be articulated, linking the teacher now to one, now to another benefactor—a parent, a town council, a provincial governor, or an emperor.

These relations had their ideological foundation in the union of doctrina and mores , the combination already found in the heroes and villains of the literary tradition. The centrifugal force of learning, tending toward personal distinction and autonomy, was balanced by the centripetal force of mores , urging conformity to established values and behavior. The meeting of the two forces and the resolution of their tensions are differently expressed in different contexts, accordingly as the grammarian is glimpsed in his professional writings or in a social setting.^[136] There is, overall, a strong normative urge to resolve the tensions by subordinating doctrina to mores .

First, good learning and good mores are assumed to be inseparable: Gellius's half-learned grammarians are not quite gentlemen; his gentleman grammarian is among the most learned men he ever knew. The union of qualities is part of the line of continuity in the classical tradition from the early to the late empire and between literary and social convention. Macrobius's praise of his idealized grammarian, "at once admirable in his doctrina and attractive in his verecundia, " recapitulates the qualities patrons desired in their dependents and imagines a man capable of finding the middle course between the extremes of professional life described in a fourth-century letter of recommendation: "You know, of course, . . . how rare is the affinity of eloquence and good character [bonum pectus ]: either verecundia diminishes an unassuming talent [modestum ingenium ] or the eloquent man gets above himself in his success."^[137]

[136] See Chaps. 4 and 5, and Chap. 6, respectively.

[137] Symm. Ep . 1.43, recommending an advocatus to Ausonius; for verecundia in an earlier letter of recommendation, see, e.g., Fronto Ad Velum imp . 2.7.7 (p. 128.11f. van den Hout): nihil isto homine officiosius est, nihil modestius, nihil verecundius . For other examples of the mores et doctrina type in the commendaticiae of Symmachus, see esp. Ep . 1.15, 79; 2.2, 16, 29, 39; 3.22; 7.58, 91; 9.2, 54. Specimens from other sources are collected by Pedersen, Late Roman Public Professionalism 30f. nn. 84, 85. Compare also Auson. Prof . 10.37-41, on the grammarian Ammonius, doctrina exiguus, / moribus implacidis: / proinde ut meritum, / famam habuit tenuem ; and esp. the editing tale of Marcellus in Prof . 18: his talent (indoles egregia ) brought him success, including a good marriage, but his bad character (pravum ingenium ) led—inevitably—to his ruin (lines 9-10; cf. Part II no. 94).

The conventions suggest that at least ideologically the grammarian's profession was not a carrière ouverte aux talents in any simple sense. Doctrina and mores might overcome the absence of high birth, but mere talent was not enough—unless "talent" be taken to include the seemingly innate qualities, on the order of naturalis verecundia , that predispose one to respectful and scrupulous behavior. Hence the first assumption slides easily into a second: doctrina and mores are not simply inseparable qualities or opposing forces in equipoise; the former is subordinate to the latter. When the emperor Julian laid it down that teachers "ought to be surpassing in their mores First of all, and then in the skills of speech," the novelty probably lay less in the sentiment than in the sectarian use to which it was presently put, in driving the Christian teachers from the schools.^[138] The sentiment surely overlaps with the Saturnalia 's pervasive belief that virtue is the prerequisite for true learning.^[139]

The emphasis on ethical qualities (especially those conducive to stability and hierarchy) as attributes equal or superior in importance to skills, which we take to be the primary qualifications for a profession, resulted from the direct contact between the professional and amateur spheres.^[140] It should be added at once, however, that this contact did not produce a broad conflict along lines clearly drawn; we might think more usefully of an interpenetration of the two spheres. In this respect it is symptomatic that, in Macrobius certainly, but already in Gellius, a chief attribute of the professional that we might assume would separate him from the dilettante is taken over by the amateur literary tradition and regarded as a moral trait, one of the attributes of the good man—his scrupulous attention to the details of his cultural tradition (what impatient modern readers of Macrobius and Gellius commonly call their "pedantry"). Conversely, Servius, the grammarian whom Macrobius idealized, can remark in the introduction to one of his technical treatises, "I met with Horace when I was at my leisure in Campania":^[141] a work that would seem to be a piece of professional writing is presented as a parergon of the scholarly leisure (otium ) affected in the West by the amateur litterateurs of the aristocracy.

Nowhere is this interpenetration more apparent than in the value set on the personal bond between teacher and pupil. This urge to intimate attachment, exemplified, as we have seen, by the follower (sectator ), is

[138] CTh 13.3.5 (17 June 362), elaborated in the subsequent directive specifically affecting the Christians, Ep . 61c Bidez.

[139] Cf. also at Chap. 1 n. 13.

[140] Passages that seem to base judgment of qualifications primarily or purely on skills are rare: see esp. Auson. Prof . 7.9-11, on the grammarian Leontius signo Lascivus; and cf. Prof . 9, on Leontius's brother, Iucundus.

[141] GL 4.468.6: Horatium, cum in Campania otiarer, excepi .

expressed with special force in the blending of the images of teacher and father, and follows from the father's responsibility for his son's education. Gellius and Macrobius dedicated their works to their sons and in so doing took their places in a long and broad tradition, putting their accumulated wealth of learning and wisdom at their sons' disposal as part of their patrimony.^[142] The literary convention, like most conventions, is a compound of actual practice and normative pressure: it reflects both the fact that a father supervised his son's education and the belief that such was the father's proper role.

The supervision might be direct: far from being a quaint and isolated figure, the elder Cato in taking personal charge of his son's lessons^[143] is part of a tradition that runs from one of the earliest references we have to , in the third century B.C. ,^[144] to the households of late antiquity.^[145] But the father's responsibility is no less emphasized when the supervision is mediated by the professional teacher. Libanius's extensive correspondence with his students' fathers^[146] and his shock that one of his enemies would canvass prospective students' mothers^[147] belong to the same world where a marriage contract stipulates a father's responsibility for his sons' liberal education,^[148] or where a student writes asking his father to visit "so that you might learn whether or not the teacher is paying attention to me"^[149] or asking his father to write "so

[142] For the dedication of works of literary scholarship to sons in late antiquity, see, e.g., Mart. Cap. De nupt . 1.2. and 9.997, 1000; Fl. Mallius Theodorus De metris, GL 6.585f.; Nonius Marcellus (cf. Part II no. 237); Ti. Claudius Donatus (date uncertain; cf. Part II no. 209); Vibius Sequester De fluminibus , dedicated to his son Virgilianus, who was probably a grammarian (cf. Part II nos. 163, 254); Ioannes Stobaeus (Phot. Bibl . cod. 167, 2.149 Henry); and probably Fl. Sosipater Charisius (cf. n. 153 below). For the tradition active in other areas, cf. Basil's dedication of his essay on Greek literature to his nephews; Claudius Marius Victor's dedication of his Comm. in Gen . to his son (Gennad. De vir. ill . 60); and Eucherius of Lyon's dedication of the Liber formularum spiritalis intelligentiae to his son.

[143] Plut. Cat. mai . 20.

[145] See, e.g., Symm. Ep . 4.20.2, 6.61; Eustrat. presb. V. Eutych . 8 (PG 86:2.2284); Callin. V. Hypat . 1. Cf. Paulin. Pell. Euchar . 60ff., the pietatis opus studiumque insigne parentum associated with his education at home; Greg. Naz. Or . 43.12; Aug. Ep . 2 12-13, esp. 13.2-3; Sidon. Apoll. Ep . 4.12.1.

[146] Cf. Petit, Étudiants 18ff., 104f., and esp. 151ff.

[147] Or . 39.4.

[149] SB 3.6262 (s.III).

that I might pay reverence to your hand, because you have educated me nobly."^[150]

Injecting himself into this deeply traditional and highly charged relationship, the teacher in turn was captured by it. In one of the most common images of the teacher in late antiquity, he assumes the role of surrogate father: he "nurtures" his students, his "children"; he is their "father in letters."^[151] The literary image—again, not merely a convention—also appears in another form. Though amateur litterateurs dedicated their works to learned friends or sons, no professional grammarian we know in late antiquity dedicated a work to his own son; dedications were to learned friends or patrons, or to pupils.^[152] A double impulse can be seen here. The professional's distinguishing mark was his stepping aside from his own role as father,^[153] but this withdrawal took the form of a transference, with the dedications to pupils reproducing the traditional pattern of family relationships. The relationship was not one-sided: it was reciprocated, for example, in a former pupil's funeral oration for his teacher or in the inscriptions recording dedications students made to teachers.^[154]

[152] For grammarians' dedications, see Part II nos. 19, 31, 47, 52, 60, 72, 73, 110, 126, 132, 136, 156; cf. nos. 188, 265. For dedications to pupils, see nos. 57, 130, 136; cf. no. 221.

[153] Thus the examples are sufficiently numerous, and the distinction sufficiently clear, that Charisius's dedication of his work to his son can add further and perhaps decisive weight to the evidence that he was an amateur and a palatine official rather than a professional grammarian: see Part II no. 200.

[154] Funeral oration: Ael. Arist. Or . 32, for Alexander of Cotyaeum; cf. Lib. Or . 1.105, for Zenobius. Dedications (a selection): CIL 8.5228 (an. 211/12, Thibilis), with 5229; IG 14.2454 (Massilia); CIL 6.9444, 9449, 9827; ILCV 721 = CIL 6.10008; IG 2 3897, with Raubitschek, "Phaidros" 99f.; IG 2 3793 (s.I); Raubitschek, "Greek Inscriptions" 248f. no. 10; IG 2 3813 (s.III med.); IDelos 1801; Robert, Collection no. 46 (Madytos, Thracian Chersonese), with ibid. pp. 56ff. for other examples; Inschr. Eph . 7:2 (IGSK 17:2) 4340; AE 1941, 141 (Prusa); MAMA 7.358 (Carbasli^[*] Yaila, eastern Phrygia). See also the comparable expressions of piety in Dracontius Rom . 1 and 3, for his teacher Felicianus (cf. Part II no. 59); the poem of Georgius for his teacher Coluthus (cf. Part II nos. 33, 63); and Priscian on his teacher Theoctistus (esp. GL 2.238.5f.). Note that the sentiment is not confined to teachers of liberal studies: see Sahin^[*] , "Neue Inschriften" 34 no. 103 (with C. P. Jones, "Two Epigrams"), a sepulchral inscription set up by an apprentice weaver to his teacher with the appropriate expressions of regard and gratitude.

Such details trace the normative model of relations between teacher and student. They are the concrete tokens of the desire for close attachment and of the belief that, in a favored phrase of Libanius, teachers act on their students' souls.^[155] To be sure, it was not the only image of the teacher,^[156] and the ideal of personal attachment was often violated in practice. But the ideal was far from inert; it prepared the way for action, to the teacher's benefit: thus Libanius asks a provincial governor to show favor () to the brother-in-law of the grammarian Calliopius, so that Calliopius and his father, who were then teaching Libanius's son, would in turn be more favorably disposed () toward their charge.^[157]

This cooptation should not be thought of as either dramatic or conspiratorial. The ruling elite did not rise up as a body to crush a nascent professional middle class built on the independence of achieved status and personal skill; it was a case of assimilation, a gradual process leaving room for upward mobility. Nor should we expect the teachers to regard the consequences as pernicious. To refer to the example just cited from Libanius: in comparable circumstances the modern academic presumably would claim to be equally shocked by the suggestion that official favor shown to one of his connections could influence professional dealings with a pupil and by the suggestion that his personal relations could justify manipulating the public administration for the sake of personal privilege. He would be inclined to find the transaction thoroughly corrupt, from the assault on his professional integrity to the undermining of one of the cornerstones of the common good, government without fear or favor. But the ancient academic would find equally shocking the suggestion that any discontinuity existed between professional relations and personal relations, between personal relations and customary privilege, between customary privilege and social good.

The arrangement Libanius sketches was not simply ordinary; it was proper. The outrage lay in its violation.^[158] One can diagnose corruption in this sort of transaction only from outside the social system in which the transaction was embedded, and with a different image of community and social good in mind. Historical remove can provide such a vantage point, allowing us to describe as "self-satisfied idiocy" the identification

[155] Cf. Ep . 337.1, 398.2, 969.1.

[156] For an opposing image, the Christian teacher as father vs. the grammarian with his cane, cf. Aug. Tract . 1, p. 449.31ff. Morin.

[157] Lib. Ep . 678 to Iulianus, praeses Euphrat . See further Chap. 6, pp. 209ff.

of the learned with the good and to regret all the consequences that flow from it.^[159] Another such perspective might be found in a contemporary culture with different roots or a different angle of vision—the culture of the Christians, to whom we now turn.