[In the following essay, Carter surveys the content, structure, style, sources, influence, textual history, and reception of Valerius's Memorable Doings and Sayings, commenting primarily on the work's stylistic limitations and the reasons for its centuries-long popularity.]

The great German nineteenth-century historian, Niebuhr, was perhaps exaggerating when he said that the Middle Ages considered Valerius Maximus ‘the most important book next to the Bible’¹ but he was certainly well-known and much read by all who counted themselves educated from the time of Charlemagne to the sixteenth century, and when, for example, the foundation statutes of Corpus Christi College, Oxford, were drawn up in 1517, Valerius was one of the four prose authors prescribed for daily lectures to first-year students.² Yet a Classics student at University today can normally expect to graduate without ever having read a word of Valerius and unaware that medieval and Renaissance Europe learnt—or thought it learnt—more about the Romans from Valerius than from any other single Latin author. This dramatic change in popularity and importance is only one of the things that makes Valerius so fascinating. He turns up in odd places even today.³ In many ways Valerius never deserved to survive, and he still obstinately refuses to die. Paradoxes like this surround an author whom modern taste rightly finds one of the most tedious and affected products of the ancient world.

He is known to us by only one work, the Facta et Dicta Memorabilia or Famous Deeds and Sayings, transmitted in nine books and containing about a thousand stories and anecdotes varying in length from a bare three or four lines to extended and elaborate versions of more than a page.⁴ They are broadly classified into ninety-odd groups or chapters, and supplied with chapter headings, not all of which are original.

Book I is devoted to religious topics of the more sensational kind familiar to readers of Herodotus, Livy and ancient historiography in general: the advantages of religious observance, the dire results of its neglect or dissimulation, auspices,⁵ omens, prodigies, dreams and finally miracula, which also embrace folk-myth wonders such as the Asiatic princess with the double row of teeth, the man who could see across the Mediterranean from Sicily to North Africa and the goats of Cephalonia that drink the wind (I.8.12ff). Book II illustrates traditional Roman customs, institutions and behaviour. In the first half come stories about marriage and divorce, politics and administration, the army, public festivals etc., with a section giving a few comparable foreign practices (II.5); the second half concentrates on those still more emotive sources of Roman self-congratulation—military discipline, the triumph, the censorship, and the perpetual aura of respect traditionally surrounding the distinguished (maiestas clarorum virorum) even in the most unlikely or deprived circumstances. In these two books and throughout, the emphasis is on the individual story and the dogged exploitation of its rhetorical, dramatic and moralising aspects. The coverage is not comprehensive, and the examples do not amount to a dispassionate account of each topic. Although his subject-matter is primarily historical Valerius was not writing history,⁶ nor was he some ancient sociologist. His work is no more than a product of the declamatory tradition, and this becomes clear as we wade through the relentless stream of stories about Character and Morality in books III-V. The capital letters convey a sense of their pompously didactic tone and Victorian earnestness.

The bulk of book III is made up of sections on bravery (the longest in the work), resolution, self-confidence and constancy, prefaced by a small collection of precocious Roman infants displaying exemplary Romanitas. In book IV comes the turn of moderation, abstinence, modesty and friendship; in book V, kindness and mercy, gratitude and ingratitude, and pietas, filial, family and national. Between these main sections are a number of shorter and less neatly defined topics—people who rose from humble origins, people who let their noble families down; fathers who loved their children, fathers who were ruthless to their children, fathers who were kind to their children when they were suspected of something, fathers who resolutely faced the deaths of their children; and a title of more immediate appeal these days, Famous People who indulged in More Outrageous Dress or Behaviour than was Traditionally Acceptable (III.6). There is no obvious or subtle unity to these books, individually speaking, and their collective content has no aesthetic pattern or logical coherence. Closely related topics or similar stories can be widely separated between books. This again is characteristic of the whole work, and books VI-VIII are even less homogeneous than the previous five, with morals, history, social phenomena and antiquarianism ranged side by side and a great deal that would be equally or more at home in books I-V. The sixteen stories On Chastity at the beginning of book VI echo those on modesty and conjugal love (IV.5 and 6); the next three chapters illustrate freedom of speech and action (cf. IV.8), ruthlessness (cf. V.8) and dignity of speech and action (cf. II.10; VI.2); the fifth chapter, justice in and out of the law-courts (cf. VIII. 1-6); and the remainder, trust in politics, trust between husband and wife, trust between slave and master (cf. III.8; IV.6; V.2) and reversals of character and fortune (cf. III.4 and 5).

Book VII is a similar pot-pourri. The first chapter, On Good Fortune, is explicitly linked to the final chapter of the previous book, and the next two—wisdom of speech and action, artfulness of speech and action—also recall chapters in book VI. After a section on military strategy and tactics (cf. II.7; III.7), the rest of book VII skips from six anecdotes about electoral defeats and ten of Dire Straits befalling Our Nation and Heroes to three chapters about wills and inheritance—last wills and testaments revoked, last wills and testaments that were not revoked when they should have been, and people who nominated unexpected heirs. These lead straight into the six legal chapters that open book VIII, and again it is a mixed bag. In modern editions the first chapter is a composite one—thirteen instances of the acquittal of the guilty, eight very much briefer cases of the condemnation of the innocent and two adjournments sine die, both involving women; this section is followed by some famous civil lawsuits, more trials figuring women, evidence obtained from tortured slaves, famous testimonies and the final recherché; quirkiness of People who committed the Same Offences for which they had Procured the Punishment of Others. At this point Valerius considered his audience's legal needs had been satisfied, or his own patience or sources were exhausted, and general diversity becomes the order of the day. Effort and Industry and The Uses of Leisure are succeeded by The Power of Oratory and The Importance of Clear Pronunciation and Apt Gesture (VIII.7-10), then two meagre sections follow on the Arts—meagre because the foreigners begin to seriously outnumber the Romans, so that the topics are dropped in favour of depressingly familiar matter, On Longevity, Love of Glory and The Rewards of Excellence (VIII.13-15).

Book IX is more like books III-V and more consistent at first. The main theme is deficiencies of character and Vice, with the first ten sections exemplifying extravagance and lust, cruelty, anger and hatred, avarice, arrogance, treachery, violence including sedition and mutiny, foolhardiness, error, vengeance. But the final five or six chapters return to the world of jumble with Dreadful Deeds and Wicked Words, Peculiar and Remarkable Deaths, Unmanly Failures to Face Death (this includes supercilious accounts of elaborate precautions against death taken by foreign tyrants—IX.3.ext.2-4), Confusions of Identity and ‘Doubles’ and a last section containing seven pretenders to noble birth and eminence, five Roman, a woman from Milan and an Asiatic tribesman.

Here the Facta et Dicta comes to an abrupt halt, without epilogue or final flourish. This is uncharacteristic of Valerius who is chronically effusive and sententious, especially at beginnings and endings, and although the work is the unique survivor of its type, it is unlikely to be complete as we now have it. In the preface to his epitome of Valerius, compiled perhaps in the fourth century, a certain Julius Paris⁷ refers to ten books of Valerius, but his epitome ends book IX with the same anecdote as the full text; what then follows as book X is a fragment of an entirely different character about Roman proper names, attributed to an earlier epitomator, C. Titius Probus. Figures and numbers are the most unreliable thing about any ancient text, and book-division was also a much more flexible thing in the ancient world: at the one place where the text refers to Valerius, the manuscripts of Aulus Gellius, for example, cite book nine for a story that appears in our book eight.⁸ Since the work has no definable plan, it is impossible to calculate the extent of any loss or know the cause. It could be anything from one or two more stories and a short epilogue, lost with the last leaf of some ancient codex, to something very much longer—one or more books, perhaps separated from the extant nine in the days of the papyrus roll and never making that crucial transfer into codex form.⁹ Further additions might have been halted by Valerius' death. It is useful to clarify the background a little: one cannot be reminded too often, particularly where antiquity is concerned, that the history of the past is ‘an enormous jig-saw with a lot of missing parts’.¹⁰

The stories themselves range from old chestnuts like Marius and the donkey, Coriolanus and his mother, and the air-borne tortoise that reputedly slew Aeschylus (I.5.5; V.4.1; IX.12.2) to the contented suicide of a nonagenarian old lady witnessed by Valerius himself on the island of Cos (II.6.8). The time-span stretches from Romulus to the Emperor Tiberius, though the frequency diminishes rapidly after Caesar's assassination and the battle of Actium. Most stories are about the mighty and exalted, but the lower orders are admitted where their exploits are sufficiently memorable (I.7.4; IV.7.5, etc.). Unless the theme is purely Roman like the military triumph or the censorship, a series of Roman examples is followed by a group of externa, chiefly Greek; these are generally fewer and shorter and form about a third of the complete total. When one views the enormous variety of topic and range of material, the appetite is whetted. Unfortunately, the dazzling possibilities of the subject-matter are smothered by Valerius' style, and to tackle the stuff in any quantity becomes an increasingly gloomy and indigestible experience. The approach is uniformly dull, monotonously turgid and oppressively forced, and variations in tone, length of narration and use of direct speech are too sporadic and slight to make any difference. The declamatory conventions of his day have little to do with it. What distinguishes Valerius from Lucan, Seneca, Petronius and Tacitus is supreme mediocrity of talent. He was no spinner of words and whatever the cultural environment a silk purse cannot be made out of a sow's ear.

But before studying his style in greater detail, something must be said about the date of the work, its more precise nature, sources and value to the ancient historian.

Valerius appears to have lived and written during the reign of Tiberius. Augustus is dead (VII.8.6; IX.15.2) and Tiberius is princeps parensque noster, a divinity still present to guard and guide in all things (I. praef.; II. praef.; V.5.3; VIII.13. init.). There is no mention of his death or events after it.¹¹ Pliny the Elder quotes Valerius in book I of the Natural Histories as a source for books VII and XXXIII: passages in both these and other books¹² could certainly derive from Valerius though not necessarily from him alone. These are the earliest extant references. The declamatory, pointed style of the work (Seneca the Elder is closest in spirit though superior in quality) is also consistent with an early first-century a.d. date. Attempts have been made to date the Facta et Dicta more precisely but scholars disagree whether Velleius Paterculus, who dedicated his Historiae Romanae to the consul of a.d. 30, used Valerius or vice versa or whether both derived the same or similar material from the same or similar sources; the corruption of Velleius' text adds to the problem. But there has been universal agreement since 1854 that Valerius wrote between a.d. 27 and 31.¹³ The evidence is all internal and none of it satisfactory. This is not the place to debate the matter in detail, but the argumentation has a wider relevance, so is worth pursuing a little here. Many a ‘dateable’ passage in ancient literature is equally suspect.

The whole case rests on a mere three passages. In the first Valerius tells us that his visit to Cos en route for Asia Minor was graced by the presence of a certain Sextus Pompeius (II.6.8); it is assumed that this is the Sextus Pompeius who was consul in a.d. 14 and proconsular governor of Asia soon after a.d. 27.¹⁴ The second is the preface to the opening chapter of book VI, Valerius' unctuously gauche invocation to Chastity and ‘her persistent guard over the nuptial couch of the Julii’ or ‘nuptial couch of Julia’ if the reading of the manuscripts is accepted.¹⁵ The argument goes that this is an indirect reference to the Empress Livia, who adopted the official name of Julia after the death of Augustus; therefore Valerius wrote the beginning of book VI before Livia's death in a.d. 29. The third passage hysterically denounces the downfall and elimination of an unnamed figure who ‘extinguished the light of trust and friendship’ and ‘endeavoured to wrest the reins of empire from the secure grasp of our beloved Father and Emperor’ (IX.11. ext. 4 fin.). Sejanus is the most obvious candidate and book IX is therefore dated closely after Sejanus' death in October a.d. 31. A neat pattern emerges, pleasing to the bureaucratic tidy mind: Valerius had got to the middle of book II soon after a.d. 27, to the end of book V by 29 and laid down his pen during the winter of 31/32.

There are objections to all these conclusions. In the passage about Sextus Pompeius Valerius does not tell us about any consulship or Asiatic proconsulship. In fact he tells us nothing about his Sextus Pompeius apart from conventional tributes to his superlative character, kindness and eloquence. It is a reasonable inference from the only other place he appears that he was Valerius' patron (IV.7. ext. 2 fin.), but here again nothing specific is said or implied about his career or status apart from his ‘ruin’,¹⁶ the main subject of the passage. Consider on the other hand the eminence of the consul of a.d. 14.¹⁷ He may well have been Ovid's patron and recipient of four of the Letters from Pontus.¹⁸ If we can trust Dio, he was a relative of Augustus,¹⁹ though the connection is obscure and may be as vague and remote as indirect kinship with Julius Caesar via his marriage with the daughter of Pompey the Great. As consul of a.d. 14, he also enjoyed a unique distinction. This being the year of Augustus' death, Pompeius and his consular colleague administered the new oath of allegiance to the Senate and people etc.²⁰ and proclaimed the accession of the monarch Valerius flatters with rhetorical regularity, assisting at the birth of what Valerius calls ‘the most felicitous era the Roman people have ever enjoyed’ (VIII.13. init.). He then secured perhaps the prize proconsular governorship—a Senate appointment, but the holder was unlikely to have been out of favour with the Emperor. Of his subsequent career nothing is known and Tacitus for example has no record that the consul of a.d. 14 was ruined or died in disgrace during Tiberius' reign. It is difficult to believe that Valerius would avoid all reference to his beloved patron's distinguished connections and career if indeed he was the consul of 14, and the straightforward insertion of an ‘omitted’ procos. towards the beginning of the story in book II is no solution to the problem. No, the most economical path is not necessarily the right one, here or anywhere. Let us admit that we do not have a record of all the people in Italy by the name of Sextus Pompeius and that proconsuls were not the only people who visited the provinces during Tiberius' reign. Of the six or so known Pompeii,²¹ the eques who made Tiberius cross on one occasion or the Pompeius who died on a conspiracy charge in the early 30s²² look much more promising than the consul of a.d. 14.

The invocation to Chastity involves a matter of basic interpretation. It is generally overlooked that Tiberius is a pervasive presence here—Valerius' supreme incarnation of Chastity.²³ The ‘persistent guard over a nuptial couch’ is a clumsy but simple allusion to the fact that he never remarried. If there is any Julia here it is the obvious one, Tiberius' former wife, the notorious and adulterous daughter of Augustus. It is true, of course, that Tiberius was her third husband, that they became estranged, that she was sent into exile in 2 b.c. for alleged profligacy, and Tiberius himself may have authorised her starvation to death as one of his first acts after accession. An invocation to Chastity that explicitly embraced such a flagrant example of the very opposite would be tasteless and inept. But Valerius is a tasteless and inept writer and provincial enough not to have known or forgotten a cause célèbre fifteen or twenty years past. It is a fallacy to assume that every Roman was omnicompetent and infallibly correct in matters of taste.²⁴ To inject Livia into the passage, ingeniously exploiting that ‘happy’ coincidence of name, is a distortion of the natural sense and hopelessly contrived. In any case Livia's fidelity to the dead memory of Augustus was a national legend, and Chastity personified in the form of her son could celebrate, honour and guard her literal or metaphorical marriage-bed irrespective of her existence or present occupancy of the bed. There is no sense in which this invocation had to be written before a.d. 29.

The third passage may well be a diatribe against Sejanus and its melodramatic namelessness and fevered intensity may reflect comparative closeness to the fall itself. If so, a collection of about a thousand separated stories totalling some 80,000 words now depends for its precise dating on this one passage and there are considerations applicable to all dateable references that need scrupulous attention. The first is the question of originality, the second of unity with the surrounding context, both affected by the fluid uncertainties of ‘publication’ and manuscript circulation in antiquity. The style of the diatribe is not conspicuously different from the rest of the work and displays all the signs of bombastic lubberliness so typical of Valerius. But this is not saying much. In the first century a.d. Italy was full of second- and third-rate declaimers like Valerius and to a large degree their styles will have been indistinguishable. Only writers of genuine talent have distinctive styles that are difficult to emulate. But even if we assume Valerius' authorship, its position is abnormal—a Roman story placed in the externa. It could easily have been jotted down and inserted long after the rest was written. A work like the Facta et Dicta invites and readily absorbs additions by author and reader alike.²⁵ The diatribe is not a reliable guide to the date of the rest of book IX, still less the whole work.

The precise date of the work's composition is not a matter of world-shattering importance. Handling familiar material in what was then a universal style as anonymous as modern officialese, the Facta et Dicta was clearly compiled under Tiberius, who is periodically addressed and flattered throughout; exacter dating is a dispensable refinement, but it does no harm to encourage scepticism even in small matters and very minor authors.

About Valerius himself we know nothing. There is no external evidence and the ‘Life’ found at the beginning of some medieval manuscripts is a medieval distillation of inferences from the work itself. The passage lamenting his patron's downfall seems to indicate a man of little status and small means, but allowances have to be made for rhetorical exaggeration.²⁶ Numerous references are made to notable Valerii from the past and their exploits, but there are no hints about ancestral connections with the author. Similarly there is nothing to indicate his birth-place or home; he could have been a Roman, an Italian or a provincial.²⁷ There are frequent stories about generals and armies, but nothing about them suggesting personal experience of military life. The ex-soldier Frontinus, relating the same or similar stories, uses simple technical terms and directer language than Valerius, but it is unsafe to conclude that Valerius never served or fought. He was obviously an educated man with some knowledge of rhetorical theory and his work was primarily of use to students of rhetoric and declamatores, but this is not proof positive that he was a practising speaker himself or taught the subject professionally. Which brings us to the question of the nature and function of the work and its background.

The Facta et Dicta is a hand-book of exempla, i.e. stories or examples of behaviour, attitudes, comments, ‘happenings’, usually with strong didactic overtones. Such illustrations added colour, variety, interest and persuasiveness to both conversation and formal oratory,²⁸ just as we enliven discussion today with items from newspapers, general reading or personal experience. History and philosophy provided the chief sources of material, especially the former. Mainly under the influence of Isocrates, Xenophon, Theopompus and other Greek men-of-letters of the fourth century b.c., the boundaries between these and oratory had become less distinct. The Romans inherited this process at an advanced stage: for Cicero, historiography was ‘an highly oratorical affair’ (opus oratorium maxime)²⁹ and oratory steeped in historical allusions. Cicero also expressed the importance of philosophy to the orator in a passage that touches on the raison d'être of exempla and demonstrates their use (Orator, 4.14-15)

We shall appreciate more fully later on what is here assumed from the outset, that philosophy is indispensable for the production of our orator, not because philosophy is the be-all and end-all but because it helps in the way that physical training helps the actor (I make this comparison because it is often highly appropriate to compare great with small). It is impossible to discourse on a variety of noble topics in a fluent, comprehensive and satisfying manner without a knowledge of philosophy: Socrates, for example, in Plato's Phaedo says that Pericles excelled other orators because he had been a pupil of Anaxagoras, the natural philosopher. According to Socrates he learnt much that was sublime and magnificent from Anaxagoras, the source of his fertile eloquence and all-important knowledge of the various emotional reactions stimulated by each type of oratory. Demosthenes may be considered another example: we can appreciate from his letters how frequently he sat at Plato's feet. …

Cicero's whole output—speeches, philosophical and rhetorical works, letters—is full of exempla, from one-word references to detailed vignettes. A corpus of the more familiar or striking exempla naturally accumulated through time in all forms of oratory and in the increasing numbers of histories, biographies and autobiographies that provided their first breeding-ground. Further assisted by the mammoth encyclopaedic labours of scholars and librarians, the standardisation of educational rhetoric and spread of the rhetorical schools,³⁰ the collection of exempla in more readily accessible form as reference books was an inevitable development. Cornelius Nepos, who died in 24 b.c., produced one such collection and Aulus Gellius refers to a story that appeared in its fifth book.³¹ Augustus' freedman and Palatine librarian, Hyginus, produced another and Valerius cites the Collecta of a certain Pomponius Rufus (IV.4. init.). None of these have survived, and there were doubtless many others of which we have no record; some schools of rhetoric are likely to have made their own hand-books or supplemented the editions in general circulation. Similar collections were also available, like Varro's Human and Divine Antiquities in 41 books, Libri Logistoricon (dialogues with titles like Marius on Fortune) in 76, his 15 books of portrait studies of famous Greeks and Romans called Hebdomades, Nepos' De Viris Illustribus (known to contain at least 16 books and part of which—Famous Foreign Generals—has survived), Hyginus' De Vita Rebusque Illustrium Virorum or the Libri Rerum Memoria Dignarum or Handbook of Memorabilia by Verrius Flaccus, who tutored Augustus' grandsons. More specialist works on religion or law will also have been pressed into service. By the early first century a.d. there was an abundance of material to draw on, more extensive than we know, and, apart from the fragment of Nepos, Valerius' Facta et Dicta is the sole survivor.

In the preface to book I Valerius describes its composition and purpose in vague, flowery terms:

The history of Rome and of foreign nations supplies us with many deeds and sayings worthy of remembrance, but they are distributed and spread at large among a host of other writers and cannot be apprehended quickly and concisely, so I have decided to make a compendium of selections from illustrious authorities³² so that those who want illustrations³³ may be spared the labour of long research. But I have resisted any urge to be comprehensive. …³⁴

In other words he has compiled the ancient equivalent of a Dictionary of Quotations. The content was universally appealing, especially to students of rhetoric and practising orators, but we cannot tell whether it was the personal manual of a practising rhetorician or the private product of Valerius' leisure or retirement—a sort of commonplace book—which he or others thought worthy of a wider audience and tidied up for publication.

He is normally silent about his sources, unlike Cicero in the first book of De Divinatione for example, which at times is a continuous series of exempla. Coelius Antipater is cited for C. Gracchus' dream about his brother (I.7.6) and Livy for the vast serpent that terrified Atilius Regulus' army in Africa in 255 b.c. (I.8. ext. 19). These are the only two references in book I. There are none at all in book II. ‘Reliable authorities including M. Varro’ attest the otherwise incredible military exploits of L. Siccius Dentatus (III.2.24) and in book IV we meet Munatius Rufus' biography of Cato, Pomponius Rufus' Collecta and M. Scaurus' autobiography (IV.3.2; 4. praef.; 4.11). Thus they trickle on at infrequent intervals. In book VIII the number suddenly rises with Cato's Origines, Isocrates' Panathenaikos and Cicero's Pro Gallio (VIII.1.2; 7. ext. 9; 10.3) and an unparalleled outburst at the end of VIII.13—Cicero's De Senectute, Aristoxenus the Musician, Asinius Pollio, Herodotus, Ctesias and Theopompus, Hellanicus and Damastes, Alexander Polyhistor and Xenophon of Lampsachus' Periplous tumbling out, sometimes two at a time. This is almost the sum total in the whole work,³⁵ but only the most gullible of ancient or modern readers will accept the situation at face value. Livy is mentioned the once, Cicero as a source only twice and Pompeius Trogus not at all, but these are almost certainly the ultimate sources for the bulk of the work. A comparison of De Divinatione I and book I of the Facta et Dicta illustrates the deceptiveness of appearances. Gracchus' dream appears in the Cicero, complete with reference to Coelius Antipater (I.26.56) and flanked by the famous story of the rustic at the plebeian games, Simonides' dream about the corpse on the sea-shore and the two Arcadians at Megara (I.26.55; 27.56 and 57); Valerius includes the same four stories (I.7.4, 6, ext. 3 and ext. 10) along with twenty others also found in the De Divinatione. Valerius could have taken all twenty-four from a copy of the Cicero, but the work is never mentioned.

Source-hunting is a favourite scholarly pursuit and Valerius has received his share of attention.³⁶ But two factors are commonly ignored or glossed over in many such enquiries—the survival of only a small proportion of what was written in the ancient world, and the misleading assumption that even the best modern texts are identical with their ancient counterparts and reliable bases for detailed comparisons. To take a simple example, Valerius includes the name of the rustic at the plebeian games (I.7.4) and this is not to be found in our texts of the De Divinatione. This is not proof of mutual independence: any ancient text of Cicero may have had the name scribbled in the margin or incorporated in the text; it was common knowledge anyway and Valerius could have included it without thinking. We cannot be positive that Valerius originally supplied the name in his version for the same reasons, and his modern editor has to decide between Latinius (the reading of the earliest extant manuscripts of the complete text, both of the ninth century) and T. Latinius (the reading of the ninth-century manuscript of the late Empire epitome of Julius Paris). A minor example, but the problem it illustrates is a major one. The loss of so much ancient literature also makes the whole business of source-hunting and the tracing of influences a more delicate operation than many of its practitioners seem to realise, and an account of their minute labours often excites a suspicious wonderment that such things should be. It is certainly not to be supposed that Valerius used only the sources he names, nor that he did all the donkey-work himself, or any of it.³⁷ He had his own epitomators later and there is a strong likelihood that he plundered Hyginus or Verrius Flaccus or some other lost collection. There have been detailed discussions on this point but in default of all other survivors they are academic in the worst sense. The ultimate sources are most likely to have been Cicero and Livy, and it is perhaps true that the very small proportion of Augustan and still fewer Tiberian anecdotes reflects the limits of a single direct source of late Republican or early Augustan date. The lack of contemporary or more original material in the Facta et Dicta is not to be construed as modest lack of self-confidence or political caution on Valerius' part.³⁸ Thus, recent campaigns, victories, heroic exploits and mutinies are ignored in favour of Actium and its predecessors; Marius and Sulla, Hannibal and Scipio appear time and again, but there are no Imperial generals or legates, no Tacfarinas, no Maroboduus, no Arminius, and myriad opportunities to include Livia and Sejanus under a variety of headings are not taken. Even stories of doubtful or unknown date (e.g. VIII.4 on tortured slaves) are much more likely to be Republican than Imperial, and the six different Calpurnii Pisones all belong to the previous century, with no reference to the Younger Germanicus' watch-dog, the notorious governor of Syria.

It can be understood from this that Valerius is of little value to the historian interested in the people and events he mentions. He is at best a second-hand source for everything earlier than Actium and the capture of Alexandria, and this leaves very little. His rhetorical elaboration and clumsy obscurities of style also diminish his reliability and accuracy. Where he conflicts with earlier authors or his testimony is unsupported, he is suspect. The errors and inconsistencies have been diligently hunted down—another favourite scholarly pursuit. Thus in book I Tanaquil is the wife of Ancus Martius, not Tarquinius Priscus, and the Athenians capture Aristomenes of Messana, not the Spartans (I.6.1; 8. ext. 15). Glabrio, father and son, are confused (II.5.1) and at III.2.20 Hannibal is found besieging the Romans in Capua whereas, as Kempf succinctly observed, ‘the Romans were not in control of Capua at that time, nor did Hannibal ever besiege that city, nor in fact did the events described happen to Hannibal at all: it was Hanno.’³⁹ Three if not four generations of Scipionic achievements are concentrated in the single person of Scipio Nasica (VII.5.2) and Hannibal is again given the credit for the defeat of the brothers Scipio in Spain (VIII.15.11). In the same book Euclid the geometer is confused with Euclid the Megarian philosopher (VIII.12. ext. 1) in a story that Plutarch tells of Eudoxos of Cnidus and Helicon of Cyzicus.⁴⁰ This is a brief selection. Kempf devotes seven indignant pages to the topic,⁴¹ waxing particularly wroth over internal contradictions: in book I ‘we observe’ (nunc … cernimus) the temple of Juno Moneta on the Aventine (I.8.3) but later on ‘we see’ (nunc … videmus) the same well-known building correctly located on the Capitol (VI.3.1). There was a time when passages like this were used to suggest multiple authorship but the indisputable singleness of Cervantes, Thackeray and Tolstoy, all equipped with inconsistencies of this type, has made such arguments less fashionable today.

We have no means of telling which mistakes and confusions in the Facta et Dicta are the fault of scribes and misguided annotators of manuscripts, and which are Valerius' or the fault of his sources. Some mistakes can be corrected by a simple change of letter or word, others by the short insertion of a phrase, but economy of means will not distinguish scribal errors from more original ones. All manner of fates might have befallen the text in that gulf between Valerius himself and our earliest manuscripts of the work, and if a Roman cursive manuscript figured somewhere in the lost stages of the transmission the chances of error multiply ten-fold. The other late Empire epitome of Valerius by Januarius Nepotianus mangles things still more and also gives stories that are not in our texts of the Facta et Dicta,⁴² a powerful reminder of the uncertainties of the ancient scene.

But if Valerius has little value to the historian of events, he has something to contribute to the historian of ideas, providing evidence of a first-century a.d. Roman attitude to the past, the manner and quality of that understanding and the uses to which it was put. But the contribution is slight, adding little to what is already self-evident from the speeches of Cicero, the monographs of Sallust, the extant works of the elder and younger Seneca and the titles of the lost works of Varro. However huge the medieval popularity of the Facta et Dicta, the survival of only one more book of Livy or Petronius would be a far greater treasure.

Valerius' style is perhaps the aspect of chief interest today. Disparaging comments have already been made in passing; a closer examination is necessary and proper.

In many respects, and this includes similarity of cultural background, Valerius gives the impression of some ancient Polonius. When Hamlet's mother impatiently requests ‘more matter with less art’, Polonius protests he uses no art at all and immediately proceeds to play with words and antitheses worse than ever before (Hamlet, II.2.97-105):

That he is mad, 'tis true; 'tis true, 'tis pity;
And pity 'tis 'tis true: a foolish figure;
But farewell it, for I will use no art.
Mad let us grant him then; and now remains
That we find out the cause of this effect,
Or rather say the cause of this defect,
For this effect defective comes by cause;
Thus it remains and the remainder thus.
Perpend.

If this obsession with rhetorical figures and verbal relationships is laced with the extravagant crudity of the Players' speeches about rugged Pyrrhus and the mobled queen, the resulting mixture is very like Valerius' Latin. Here is a typical short exemplum on the resolution of Cato the Younger (III.2.14):

tui quoque clarissimi excessus, Cato, Utica monumentum est, in qua ex fortissimis vulneribus tuis plus gloriae quam sanguinis manavit. siquidem constantissime in gladium incumbendo magnum hominibus documentum dedisti quanto potior esse debeat probis dignitas sine vita quam vita sine dignitate.

(Of your most illustrious demise, Cato, Utica is the memorial: in that city, from those most heroic wounds of yours, more glory than blood flowed. Truly, most resolutely falling on your sword, you provided powerful proof to Man that dignity without life is much more morally desirable to decent men than life without dignity.)

Two things stand out—the concentration of artifice and the general clumsiness, both of which make the ‘art’ more evident than the ‘matter’. The apostrophe of Cato, the personification of Utica,⁴³ the repetition of superlatives and the other cumbersome sound patterns (especially monumentum/documentum), the two pretentious antitheses (the first unnaturally forced, the second a cliché, flourished like a conjuror's rabbit at the climax of the show and having all that animal's wearisome predictability)—it is far, far too much for a story only two sentences long and starting, so to speak, cold. It lacks tact, taste, discrimination and grace: Cato's heroic suicide is lost in the welter, form is elaborated at the expense of content. The contrast with Ciceronian rhetoric is striking. There, effects are prepared for and poised, and, with the greater art that conceals art, the most artificial forms of expression seem to grow from that magnificently versatile and fluid command of vocabulary and sentence-structure with the organic inevitability of a tree. The same formal ‘devices’—the word is appropriate in Valerius—mechanically and excessively applied irrespective of subject, soon render the Facta et Dicta tedious, and their obtrusiveness is accentuated by the simplistic sentence-structure. It is no defence to say that Valerius' style is ‘pointed’ or ‘Silver’, as though a descriptive label explained and justified all. Tacitus and the younger Seneca are ‘pointed’ and ‘Silver’ in the same general sense, but even the latter's failings show a degree of linguistic and literary competence unattained by Valerius. In his monumental study of ancient style, Norden opens his very brief discussion of Valerius by describing him as ‘the first of a long series of Latin writers whose intolerable artificiality drives one to despair’ and refuses to examine what he calls his ‘repulsive’ style in detail.⁴⁴ These are not out-dated or eccentric judgments. The Facta et Dicta is an excellent example of the unsuccessful practice of unexceptionable theory.

Comparative analysis is one of the most immediate ways of appreciating differences of style and their effectiveness. Here are three passages, one by Valerius, all telling the same story. Since style demands the detailed study of originals, translations are relegated to the notes to avoid confusion.

The first version is a straight abbreviation of the second, the main features of the story preserved in the same words but rhetorical inessentials pared down or omitted. Away go the ponderous dramatics and stumbling syntax of B's first sentence (‘The next dream, although it is a trifle longer, nevertheless, because of its extreme distinctness, secures its non-omission’) and the obvious padding in line 7 that expands subveniret (‘for by his swift advent he could be extracted from the danger that threatened’). The four places where simpler, shorter Latin replaces B's studied, self-indulgent wordiness are best compared side-by-side:

There is less clutter about A and its preservation of B's pedestrian sentence-structure is in keeping with its modesty of style. In B the same basic sentence-structure conflicts with the outcrops of ambitious phraseology and the mixed vocabulary. Even minor changes of verb contribute to A's lighter tone: ‘appeared’ or ‘arose’ (obortus), ‘covered’ (adopertum) and ‘took’ (duxit) replacing B's more high-flown ‘presented itself’ (oblatus), ‘buried’ or ‘overwhelmed’ (coopertum) and ‘delivered’ (perduxit). B's style is restless, unsure, confusing.

The world of version C is accomplished and assured. The almost colloquial simplicity and understatement of the opening clause (‘Here is the story of another very clear dream’) gives way to the pleasingly varied and unobtrusive symmetries of the rest of the sentence (‘These two friends from Arcadia were travelling together and had reached Megara, where the one lodged with an innkeeper, the other at a friend's’). It makes the opening of B seem primitively artificial. C's narrative then unfolds in a beautifully orchestrated series of clauses and sentences that constantly vary in length and type, the complex balancing the simple with sophisticated fluency and ease.⁴⁹ The central antithesis (lines 10/11) is clean, quiet and natural, principally because the flow of sound and the rhythm of the words, instinctively sensed by the writer, echo and respond (sibi vivo … subvenisset, mortem suam ne inultam …); there is nothing like this wedding of sound, rhythm and sense in B's antithesis (lines 13/14) and A not surprisingly dispensed with it. The climax in C is masterly (lines 13/19). The foundations are firmly established with the single incisive petere and the transparent simplicity of its dependent clauses, a calculated throw-away before the rich excitements that follow. The climax proper shows a fine eye for dramatic detail (the repetition of mane, the mention of the peasant-carter, the human touch about asking what was in the cart, the exquisite physical precision of erutum). There is a fine sense of timing in the placing and length of clauses for maximum suspense and impact, particularly in the short sharp succession of clauses at the dénouement, where symmetry of construction is varied with great subtlety⁵⁰—a tricolon indeed, but not foisted on the context with frigid disregard for harmony and propriety: here rhetorical precept is vibrant and fully functional. With the ground so fully prepared, even the final rhetorical question can withstand its basic conventionality.

Enough has been said to show the main differences in quality and competence. The worst by far is the middle one—Valerius (I.7. ext. 10). The abbreviated version is the work of his epitomator, Julius Paris. The final one is Cicero (De Divinatione I.27.57), Valerius' direct or indirect source.

Antitheses and sententiae, especially at the beginning and end of anecdotes, personification, apostrophe, rhetorical questions and exclamations, incessant abstraction of the personal or concrete—their interplay has a fascinating awfulness in Valerius.⁵¹ He exhibits the faults typical of poor and affected writing—miscalculated effects; over-explanation; words or phrases repeated several times and then disappearing for long stretches of text; over-elaboration and pretentious diction (tumor); ugly rhythms and sound-effects (e.g. sed sacraria aedificanda sacrificiaque facienda tribuit VI.5.1c), mixed with a high proportion of slavishly Ciceronian clausulae; clarity inelegantly maintained by the proliferation of is, ille, suus, etc., and the simple stacking of clauses in series, joined by et or -que (parataxis) instead of competent subordination and periodicity, or Tacitean marvels of compression or ellipse. Sensitivity to metaphor also separates the sheep from the goats. For much of the time there is instructively little metaphor in Valerius, then comes a harsh and violent concentration. The youth of Manlius Torquatus was ‘drenched in the cloud-burst of Fortune's contempt’ so that the glory of his old age might ‘shine forth all the more radiantly’ after the storm (VI.9.1). Parental love is invited to ‘unfurl the sails of peaceful and pious affection, voyage with wholesome breeze and import its dowry of sweetness’ (V.7. init.). There is an arguable dramatic effectiveness to the most famous mixed metaphor in English literature—Hamlet's ‘take up arms against a sea of troubles’—but what are we to make of the conglomeration of images in the last paragraph of book VI (VI.9. ext. 7)?

Human prosperity is a very fragile and perishable commodity, comparable to babies' rattles.⁵² It floods in suddenly, instantly ebbs away. Lacking the support of firm roots, in no place, in no person does it stand steadfast, but hither and thither blown by Fortune's most uncertain blast, those it has raised on high and then forsaken, impoverishing them with unforeseen recoil, these it piteously submerges in the depths of disaster.

In Odes III.29 Horace deliberately adds image to changing image, exploiting the grotesque humour that results and constantly shifting the metaphor's objective focus; Valerius is like the earnest Babu reporting his mother's death: ‘Regret to inform you, the hand that rocked the cradle has kicked the bucket.’⁵³

There is also some remarkably childish verbal play. Manlius Capitolinus, who precipitated the rout of the Senones (praecipites agebas Senones) is flung precipitately headlong (praecipitatus) over the Tarpeian rock (VI.3.1a).⁵⁴ Cleanthes laboriously draws the waters of wisdom from the well of philosophy (laboriose haurientem … sapientiam) having sustained the poverty of his youth as a night-time puller of buckets from real wells (adulescentem quaestu extrahendae aquae nocturno tempore inopiam tuam sustentantem VIII.7.ext.11). The man who claimed to be Octavia's real son was ‘conveyed towards the highest ranks of audacity by the bellying sails of presumption until Augustus ordered him to be fixed to the oar of a state warship’ (IX.15.2). In the bitterness of his ‘fall’, Q. Caepio ‘outruns’ Crassus (Crassum casus acerbitate Q.Caepio praecucurrit VI.9.13). As deliberate bad jokes they have some merit: as serious attempts at fine writing and accomplished style, they powerfully suggest the reverse.

Transitions between exempla are stereotyped and mechanical. The contrast with Ovid's Metamorphoses in this respect is dramatic. Equally wearisome is the regularity of inquit after the first or second word of direct speech. What finally could be more sterile than the repetitions, triple tricola, pairs of pairs and noisy climax of his oratorical reflections on the life of Marius (VI.9.14)?

ex illo Mario tam humili Arpini, tam ignobili Romae, tam fastidiendo candidato ille Marius evasit, qui Africam subegit, qui Iugurtham regem ante currum egit, qui Teutonorum Cimbrorumque exercitus delevit, cuius bina tropaea in urbe spectantur cuius septem in fastis consulatus leguntur, cui post exilium consulem creari proscriptoque facere proscriptionem contigit. quid huius condicione inconstantius aut mutabilius? quem si inter miseros posueris, miserrimus, si inter felices felicissimus reperietur.⁵⁵

Or the banality of Alcibiades' double fate a page later (VI.9. ext. 4)?

nam Alcibiaden quasi duae fortunae partitae sunt; altera quae ei nobilitatem eximiam, abundantes divitias, formam praestantissimam, favorem civium propensum, summa imperia, praecipuas potentiae vires, flagrantissimum ingenium assignaret; altera quae damnationem, exilium, venditionem bonorum, inopiam, odium patriae, violentam mortem infligeret. nec aut haec aut illa universa, sed varie perplexa, freto atque aestui similia.⁵⁶

Valerius then is a declaimer of the worst breed. He has his place in the history of Latin declamation, but it is easy to understand why no full-scale analysis of his style and its declamatory characteristics has yet been undertaken. ‘Every deed and every saying sprinkled with sesame and poppy-seed’⁵⁷—an apt description of his Facta et Dicta.

Finally there is the question of influence, difficult in the case of a reference book: who today acknowledges use of Roget, Pears' Cyclopaedia or a dictionary? Valerius may have been known and regularly used by rhetoricians in antiquity but Tacitus and Quintilian do not mention him and there is no detectable influence in what survives of their work. The references by Pliny the Elder, Aulus Gellius, Plutarch and Priscian amount to little:⁵⁸ Valerius is a very minor source in each case and possibly indirect. The mysterious author of the fourth book of Frontinus' Strategemata was much more heavily in his debt,⁵⁹ or was using some lost source which both he and Valerius were following closely: nothing is directly acknowledged. In the early fourth century the Christian apologist and teacher of rhetoric, Lactantius, also seems to have used Valerius for his Divinae Institutiones, again without acknowledgment. The two epitomes of Valerius by Julius Paris and Nepotianus have been quoted as evidence of Valerius' popularity in the late Empire,⁶⁰ but the accidents of survival are no more to be equated with popularity than with merit or significance—the loss of three-quarters of Livy alone shows that, or Catullus' last-minute rescue in the Renaissance. The destruction of only three more Latin manuscripts any time before the ninth century would have consigned all knowledge of the two epitomators to oblivion and made Valerius himself as shadowy a figure as Deculo, Sebosus or Turranius Gracilis.

His influence in later times is a very different matter, much more fully documented.⁶¹ It stems from the slenderest of beginnings, as with all extant Latin literature—in this case from some lost copy or copies made in the eighth or early ninth century in a monastery that was probably in north-east France or north-west Germany. In the early ninth century Valerius accordingly came to the notice—among others—of that indefatigable Carolingian humanist and scholar, Lupus of Ferrières. His copy of the Facta et Dicta with his own corrections and emendations survives as one of the two oldest manuscripts available today,⁶² and unlike its twin it bequeathed a rich inheritance in several ways. From his manuscript Lupus expounded Valerius to his pupils and the extracts taken down by the best of them, Heiric of Auxerre, have also survived in several versions, along with an index to Valerius compiled in turn by Heiric's pupil, Remigius. It was later descendants of Lupus' Valerius which supplied the libraries of the newly established Cistercian houses in the early twelfth century and which brought Valerius to England for the first time, most probably carried across the Channel from Bec or Caen to Canterbury by Lanfranc or Anselm. The preface to William of Malmesbury's Polyhistor is the earliest surviving reference to Valerius in England, although ‘historians’ are specifically excluded from quotation in the text proper. William's Gesta Regum and Gesta Pontificum, however, may owe something to the Facta et Dicta as well as to Suetonius' Lives of the Caesars.

Of much greater significance is the interest in Valerius shown by John of Salisbury, Becket's greatest friend, ally and companion-in-exile, ‘the most accomplished scholar and stylist of his age, … the Erasmus, the Johnson of the twelfth century’.⁶³ Composed c. 1159 and dedicated to Becket (the original, superbly written presentation copy is happily still in England), John of Salisbury's Policraticus or Art of Statesmanship quotes liberally from Valerius, frequently verbatim for lines at a time. His actual copy of Valerius, fathered way back by Lupus', was left to the library of Chartres cathedral where John died as bishop in 1180. Becket's copy, doubtless procured for him by his friend, went with the rest of Becket's books to the library of that ‘medieval All Souls College’,⁶⁴ Christ Church, Canterbury. This was patronage of the highest, most fertile order and there is evidence strongly suggesting that John of Salisbury, like Lupus before him, corrected and emended his text of Valerius and is responsible for the curiously distinctive ‘edition’ of the Facta et Dicta circulating in England and northern Europe from the late twelfth century.

There are close threads binding Valerius, Lupus of Ferrières and John of Salisbury, but they are only part of a larger tapestry, an example of what was going on mutatis mutandis over the whole of Continental Europe. Hardly a monastery will have lacked a Valerius, judging by the regularity of his appearance in extant medieval library catalogues. The spread of the new secular institutions, the Universities, from the twelfth century onwards opened an outlet that was to become a main-stream channel. Valerius was a god-send to a world substantially deprived of Cicero, almost wholly ignorant of Livy and unable to read Greek. The tastes and critical standards indicated by the ornate, highly flavoured but often harsh and fractured Latin of medieval Europe also help to explain the popularity of a Valerius and a Seneca. Lacking competition, offering a Reader's Digest short-cut to the history of Rome, with much approved moralising en route, the Facta et Dicta was a guaranteed success. It was one of Vincent of Beauvais' main sources for the Speculum Maius, and at least once it was put into verse, of sorts.⁶⁵

Its heyday came in the fourteenth and fifteenth centuries. The sheer quantity of surviving Renaissance manuscripts—more than of any other ancient prose author, the industrious production of commentaries and epitomes, his direct and indirect influence on works like Petrarch's De Viris Illustribus, the early appearance of vernacular translations (by Heinrich Mügeln in German, by Simon de Hesdin and Nicholas de Gonesse in French, by Antoni Canals in Catalan—all in the fourteenth century; there was also an early Castilian version), these are various guides to his pervasive influence. Or we can take the single figure of Petrarch as a microcosmic probe.⁶⁶ Valerius heads the list of historians in both accounts of his favourite books and was the model for his Libri Rerum Memorandarum or Book of Memorabilia. References to the Facta et Dicta are frequent in his correspondence and marginalia. In Paris he studied under Dionigi di San Sepolcro, who composed the most popular Renaissance commentary on Valerius between 1327 and his death in 1342. Another of his teachers, Luca da Penna, also composed a commentary. There is a long list of Petrarch's friends and acquaintances, led by Boccaccio and Coluccio Salutati,⁶⁷ who are known to have read or quoted or owned the Facta et Dicta and it includes three more authors of commentaries or epitomes—Giovanni d'Andrea, Benvenuto Rambaldi da Imola and Giovanni Cavallini. Copies of Valerius were de rigueur in the libraries of the great patrons and all scholars, and if these and Petrarch represent the pyramid peak of his influence, somewhere amid the pyramid's bulk are those ordinary readers normally obliterated by history, the record of whose very existence has hung almost perversely on ownership notes in manuscripts of Valerius Maximus—people like the French lawyer, Pierre Gontier, Johannes Vries of Amsterdam, the wife of Mucilino of Forli, Nicholas of Spoleto, or the Thomas Knyvett who scribbled his name all over the massive volume that cost him 8/4d.⁶⁸

Valerius was quick to appear in print after the development of movable type and the printing press in the second half of the fifteenth century—another index of significance. The first edition came from Mentelin's press in Strasbourg most probably in the summer of 1470, followed by an entirely independent edition the following year from Peter Schoyffer's press at Mainz, which became the basis of the first Paris edition in 1475. In the meantime Wendelin of Speier had published the first Italian edition at Venice in 1471, thereafter modified and reprinted some eighteen times before the end of the century at the various presses that quickly sprang up in every major city in northern Italy.

For another hundred years or so Valerius' popularity continued unabated. Pighius' edition, for example, first published in 1567, was reprinted nearly twenty times before 1650 and a person of Montaigne's significance still uses and refers to him.⁶⁹ But the Renaissance rediscovery of the great masters of Classical prose, the enthusiasm for all things Greek that developed after Boccaccio's reintroduction of the language into Western Europe and the increasing sophistication of cultural standards began to make their mark at the expense of the second-rate. The Facta et Dicta fell a natural victim. By the seventeenth and eighteenth centuries there was better and more attractive meat in the market—and Livy was frankly easier to read. Even scholars began to shop elsewhere, and the editors of the eighteenth-century variorum edition and nineteenth-century Teubner were comparatively little men. Valerius was published in the popular Bipontine and Delphin series in 1783 and 1819, but the standard text today is the Teubner of 1888. There is no Bohn translation, no Oxford Text, no Budé, no Loeb, no version at all in English since 1678.⁷⁰ These are eloquent expressions of his contemporary status. Valerius has had a very good run for his money, much better than he deserved. But there is no prospect of a dramatic return to favour. If he still has something of a future on examination papers in Latin Unseens, this, as Valerius himself would have put it, is the last twist of the knife buried in the heart of his present obscurity.

Notes

1. Quoted by Wight Duff, A Literary History of Rome in the Silver Age (edn 3, London, 1964), p. 59 and n.2. Augustine surely has the better claim.

2. M. L. Clarke, Classical Education in Britain 1500-1900 (Cambridge 1959), p. 22.

3. E.g. in the opening chapter of Steinbeck's novel The Winter of Our Discontent (1961), where the looseness of reference reflects the character of the speaker.

4. Short: II.6. ext. 13; V.3.2d and g; IX.15.3. Long: I.8.2 and 6; II.4.5; III.2.23b.

5. The abbreviated text of one or both epitomes (see n.7) is usually supplied between I.1. ext. 4 and I.4. ext. 1 missing in all manuscripts.

6. Understandably Valerius was thought of as a historian in the Middle Ages and the myth is still current (cf. Oxford Classical Dictionary, edn 2, 1970, p. 1,106).

7. See P-W, 10.686-9. The complete text of the Paris epitome is in the Teubner Valerius (ed. Kempf, 1888) pp. 473-587. The other epitome, by Januarius Nepotianus (see P-W, 9.696), which is possibly later in date, dispenses with book-division and covers only the first two books and part of the third (text in Kempf's Teubner, pp. 592-624).

8. Gellius, XII.7.8.

9. Cf. L. D. Reynolds and N. G. Wilson, Scribes and Scholars (Oxford, 1968), pp. 30-1.

10. Cf. E. H. Carr, What is History? (Penguin edn, 1964), p. 13.

11. Cf. n.22, the remarks on Seneca, Tranq., XI.10.

12. E.g. Pliny, N.H., VIII. 45 (70) and Val. Max., VIII. 1.8.

13. P-W (series 2), 8.i.90-93. The case was first argued by Kempf in his editio maior (Berlin, 1854).

14. Kempf, ed. mai., pp. 3-6.

15. tu … Iuliae ‹gentis› genialem torum assidua statione celebras. Pighius (edn 1567) first inserted gentis, adopted by most editors since. But see n.24.

16. iactura. Valerius' rhetoric is so woolly that it is impossible to decide which of the three chief metaphorical meanings (bankruptcy, disfavour, death) is intended.

17. PIR, 3 P, 450.

18. Epp. ex Ponto, IV.1, 4, 5 and 15.

19. Dio, 56.29.5.

20. Tacitus, Ann., I.7.

21. PIR (loc. cit.), 440, 443, 449, 450, 463, 471, 473, 480.

22. Suetonius, Tib., 57; Tacitus, Ann., VI.14. They are not necessarily the same man of course. Seneca, Tranq., XI.10 mentions a Pompeius starved to death by Caligula. It argues a desperate mental economy to pluck the praenomen Sextus from thin air, identify Seneca's Pompeius with Valerius' and the consul of a.d. 14 and advance the publication date of book IV of the Facta et Dicta to the beginning of Claudius' reign (cf. the garbled account by P. Constant in his edition of Valerius (Paris, 1935), p. ii, where books II and IV are confused). Syme ‘suspects’ the identification (AJP [American Journal of Philology], 1958, p. 21, n.14).

23. Hence the word-order at the beginning of the invocation (unde te virorum pariter ac feminarum praecipuum firmamentum …) and the imperialistic military imagery (assidua statione, celebras, praesidio, insignia, munita, numinis respectu).

24. The emendation Iuliae gentis is the unnecessary product of this false assumption. In view of gossip about Tiberius' private life, his invocation as the personification of Chastity is equally tasteless if Valerius was writing in the late 20s or 30s.

25. Cf. the remarks on the Nepotianus epitome (p. 39 above and n.42). The speech by Coriolanus' mother in Livy (II.40.5-8) is interpolated in the text of Valerius' exemplum (V.4.1) in certain late manuscripts.

26. IV.7. ext. 2 fin. Despite P-W (loc. cit.) the phrase parvulos census nostros (IV.4.11) is a human generalisation, not a description of Valerius' personal means.

27. References implying familiarity with Rome and its monuments are not proof he was a Roman or knew Rome well. A tomb at Tarentum is treated with equal familiarity (IV.6.3) and if the text is accurate he does not know the geography of the Janiculum (I.1.10).

28. Cf. Quintilian's discussion of exempla, Inst. Orat., V.11.

29. Leg., 1.2.5; cf. De Oratore, II. 51-2.

30. See part 3 of H. I. Marrou, A History of Classical Education in Antiquity (edn 3, transl., London, 1956; now in Mentor paperback); also S. F. Bonner, Roman Declamation (Liverpool, 1949).

31. Gellius, VI.18.11.

32. ab illustribus electa auctoribus digerere constitui. The ambiguity is possibly deliberate since electa could refer to already existing collections.

33. documenta, a somewhat elevated synonym of exempla here.

34. The preface to the Julius Paris epitome contains a simpler, clearer statement: ‘A supply of exempla is of course necessary for orators and declaimers so I have made a one-volume abbreviation of Valerius Maximus' Facta et Dicta.’

35. There is a full list in C. Elschner, Quaestiones Valerianae (diss., Berlin, 1864), pp. 32-42.

36. Cf. P-W (loc. cit.), 102-14, the best summary.

37. The same applies to Cornelius Nepos who litters his text with ‘sources’.

38. Though there may be something in the idea that second-hand contemporary material needed time to mature and become rhetorically acceptable.

39. Kempf, ed. mai., p. 29: cf. Livy XXV. 13-14.

40. Plutarch, Moralia, 579 (De gen. Socr. 7).

41. Kempf, ed. mai. pp. 26-34.

42. See Kempf's Teubner Valerius, pp. 599, 608-10, 616-18, 623 (=Nepotianus 7.3; 9.24 and 34; 15.1 and 17-20; 21.3).

43. For one of the worst examples of personification, cf. II.10.3 fin. (the funeral bier of Aemilius Paulus, the conqueror of Macedon, spontaneously borne by Macedonian delegates in Rome at the time): ‘Twice, Paulus, Macedonia revealed your illustriousness to our noble city—with its spoils at your safe preservation, with its shoulders at your demise.’

44. E. Norden, Die antike Kunstprosa (edn 2, Berlin, 1909) i, p. 303 and for ‘das Widerliche seines Stils’, p. 304.

45. Two friends from Arcadia travelling together came to Megara; one went to a friend's, the other lodged at an innkeeper's establishment. In a dream the one enjoying his friend's hospitality saw his companion begging him to come to the rescue—the treacherous innkeeper had tricked and trapped him. Roused by this vision he leapt out of bed and went to the inn. When he saw all was quiet round the inn he sought his bed and sleep again. The same figure then appeared all wounded and sought him to come forth and at least avenge his death: butchered by the innkeeper his body was in a cart and hurriedly being taken outside the city-gate, covered with dung. This moved his young friend who forthwith rushed to the gate, discovered the crime and brought the innkeeper to capital justice.

46. Although the next dream is a little longer, it nevertheless secures inclusion because of its extreme distinctness. Two friends from Arcadia travelling together came to Megara; one went to a friend's, the other lodged at an innkeeper's establishment. In a dream the one enjoying his friend's hospitality saw his companion begging him to come to the rescue—the treacherous innkeeper had tricked and trapped him: his swift advent could extract him from the threatening danger. Roused by this vision he leapt out of bed and endeavoured to seek the inn in which he was lodging. To his friend's calamitous misfortune he condemned a most humanitarian resolution as superfluous and sought his bed and sleep again. Then the same figure presented itself before him all wounded and implored him, since he had neglected to bring his life aid, not to deny at least murder vengeance. Butchered by the innkeeper his body was in a cart and being hurriedly taken to the city-gate, buried in dung. Compelled to action by his comrade's highly persistent entreaties he forthwith rushed to the gate, apprehended the cart which had been pointed out in the dream and delivered the innkeeper to capital justice.

47. Here is the story of another very clear dream. These two friends from Arcadia were travelling together and had reached Megara, where one lodged with an innkeeper, the other at a friend's. They each had supper and retired to bed, but in the quiet of the night the one enjoying his friend's hospitality had a dream in which the other appeared and seemed to beg his assistance because the innkeeper was plotting his death. At first he was terrified, woke from his dream and got up; then he pulled himself together, decided it was nothing after all and lay back down again. When he had fallen asleep, the same figure appeared and seemed to ask him not to let his death go unavenged, since he had not rescued him when alive. His body had been thrust in a cart by the innkeeper and a pile of dung thrown on top. He urged his companion to be at the city-gate early in the morning before the cart left. This dream really roused him. Early in the morning he was there at the gate, ready and waiting for the driver, and asked him what was in the cart. The terrified peasant fled, the corpse was dug out, all revealed and the innkeeper punished. Could there be a dream of diviner inspiration than this?

48. The syntax of pestifero fato is ambiguous and has both dative and ablative force, as frequently happens with these identical forms.

49. It is all in indirect speech, the infinitives dependent on traditum at the beginning. But these are elegantly varied and the interplay of perfect and present infinitives and pluperfect subjunctives is managed with deceptive ease, the sound-patterns contributing to the effectiveness of the climax. The extent and quality of this subordination are both beyond the writer of B.

50. ‘Symmetry and consistency … are enemies of movement’ (Kenneth Clark, Civilisation (London, 1970), 12 init.). This instinctive variety and slight asymmetry are the crucial life-giving factors.

51. Cf. Nepotianus, praef. (Kempf, p. 592): ‘Valerius writes what is worth knowing but protracts what should be concise, showing off sententiae, piling on the detail and pouring out floods of verbiage. This is the reason why Valerius is perhaps known to comparatively few because involvement and delay offend the eager reader. So I shall prune his overwriting. …’

52. crepundia, a literary and theatrical image. They were among the conventional instruments of recognition and identification in tragedy and comedy (cf. Plautus, Cist. 635-6, 664-5; Rud. 1,081 ff.; Cicero, Brut., 91.313).

53. Cf. Quiller-Couch's essay ‘On Jargon’ in the Everyman edition of his Cambridge Lectures (London, 1943), p. 92. Valerius is guilty of his two main vices of jargon—‘circumlocution rather than short straight speech’ and ‘perpetually shuffling around in the fog and cotton-wool of abstract terms’ (ibid., pp. 93 and 100).

54. When Livy tells the same story, the play of thought has a mature restraint—locusque idem in uno homine et eximiae gloriae monumentum et poenae ultimae fuit (Livy 6.20.12).

55. From the Marius that was so lowly at Arpinum, so socially inferior at Rome, so disdained a candidate for office, emerged the Marius who subdued Africa, who drove King Jugurtha before his victory-chariot, who destroyed the armies of the Teutones and Cimbri, whose double trophies are on display in the city, whose seven consulships make distinctive reading in the fasti, who succeeded from exile to consulship and proscribed to proscriber. What is more inconstant or capricious than his career? He will be discovered, depending how you class him, the most wretched of the wretched or most successful of the successful.

56. Two fates, so to speak, partitioned Alcibiades. The one allotted him illustrious nobility, abundant riches, supreme physical beauty, the favouring inclination of his fellow citizens, highest ranks of office, especial reserves of power, the fieriest of intelligences; the other inflicted condemnation, exile, distraint of goods, poverty, national odium, violent death. Neither series was uniform; they were variously interwoven like the ocean deeps a-boiling.

57. omnia dicta factaque quasi papavere et sesamo sparsa (Petronius, 1).

58. Pliny, N.H., I(7) and (33); Gellius XII.7.8; Plutarch, Brutus 53, Marcellus 30; Priscian VI.1.2 (Keil, Gramm. Lat., 2.i, p. 195).

59. Cf. the Loeb edition of Frontinus, pp. xxii-vi, the most accessible summary of the work of Wachsmuth, Wölfflin et al., with the former's list of thirty-two passages common to both (p. xxii, n.3). There are little or no signs of the Facta et Dicta in books I-III of the Strategemata.

60. Schanz-Hosius, Geschichte der römischen Literatur II (edn 4, Munich, 1935), p. 591; Wight Duff (op. cit.), p. 59; Oxford Classical Dictionary (loc. cit.). Nepotianus' own comment, quoted above (n.51) is ignored.

61. For the original material in this and following paragraphs and a critical survey of previous scholarly work, cf. my dissertation on the manuscript tradition of Valerius in Cambridge University Library; cf. also the excellent book by Reynolds and Wilson (op. cit., n.9 above) and M. Manitius, Geschichte der lateinischen Literatur des Mittelalters (Munich, 1911-31), passim.

62. Berne, Stadtbibliothek 366 (=A in Kempf's editions). Its twin (=L), also of the ninth century, is in Florence. I have examined nearly 150 manuscripts and versions of Valerius and have yet to find a descendant of L.

63. David Knowles, The Evolution of Medieval Thought (London, 1962), p. 135.

64. Ibid., p. 136.

65. By an eleventh- or twelfth-century Fleury monk, Rodulfus Tortarius. His versification survives in only one manuscript (ed. M. B. Ogle, Papers and Monographs of the American Academy in Rome, 8, 1933).

66. Cf. P. de Nolhac, Pétrarque et l'Humanisme (edn 2, Paris, 1907); B. L. Ullman, Studies in the Italian Renaissance (Rome, 1955); R. Sabbadini, Le Scoperte dei Codici Latini e Greci ne' Secoli XIVe XVe (Florence, 1905-14), passim.

67. Cf. B. L. Ullman, The Humanism of Coluccio Salutati (Padua, 1963). Coluccio suggested two emendations to Valerius—Scipione Asina for Scipione Nasica at VI.9.11 and the insertion of filius eius before qui matrem Idaeam … excepit at VII.5.2; both are occasionally found in late manuscripts.

68. Owners of Vatican, Reg. Lat. 766; British Museum, Burn. 211; Vat. Lat. 1918; British Museum, Add. 11977; Cambridge University Library, Mm.II.18 (2313). The most notable Thomas Knyvett was the Westminster JP who discovered gunpowder in the cellars of Parliament in 1605 but—another case of Sextus Pompeius!—he is unlikely to have been the Knyvett of the Cambridge manuscript.

69. Cf. J. E. Sandys, A History of Classical Scholarship, vol. ii (Cambridge, 1908), p. 197, n.5. Constant gives a list of cross-references in his edition (Paris, 1935), ii, pp. 441-2.

70. By Samuel Speed. The Valpy Delphin edition was not followed by a Valpy translation in their popular Family Classics series, but I am working on a Loeb translation and new text; with the generous assistance of the Computing Laboratory at St Andrews I have also computerised the Facta et Dicta and made a complete contextual concordance on the lines of D. W. Packard's Livy concordance (Harvard, 1968). To this extent Valerius is entering the twentieth century.

Works Cited

P-W Pauly-Wissowa-Kroll, Realencyclopädie der klassischen Altertumswissenschaft (1893-).

PIR Klebs-Dessau, Prosopographia Imperii Romani (1897-8).