3

J. E. LENDON

Historians without history: Against

Roman historiography

In history, as in life critically considered, truth rests not on possibility nor on

plausibility but on probability.
Jacques Barzun and Henry F. Graff, The Modern Researcher
(3rd edn, 1977: 133)

This essay weeps at the intellectual motion this volume exemplifies: the triumph
of what now masquerades as “Roman historiography,” the academic study
of the ancient Roman historians as a discipline sundered from Roman history,
the study of what happened in ancient Rome and why. Its narrower target is
the species of scholarship about the Latin historians arising in T. P. Wiseman’s
Clio’s Cosmetics of 1979 and A. J. Woodman’s Rhetoric in Classical Historio-
graphy of 1988.1 This writing, grown considerable in the late 1990s and
the current decade as the founders’ stars attracted satellites, studies the Latin
historians as literature. In the hunt for the historian’s artistry or ideas, his
concern with historical material – that body of “what happened in the past”
that the historian was trying to convey – is either argued away or passed over.
The Latin historian is constrained to become – depending on modern whimsy –
a rhetorician, a dramatist, a novelist, or, in the late-summer bloom of academic
narcissism, a postmodern literary critic. What the Latin historian is not allowed
to be is what he thought he chiefly was: a teller of true tales about the past.
For the contemporary historian of Rome the conclusion of this scholarship,
which Wiseman and Woodman argued in detail, is that the historical content
of the writings of ancient historians of Rome is far lower than previously
realized: that ancient historians of all periods happily invented material for
their narrative and, when writing of distant ages (such as early Rome), they

1
In the movement’s manifesto, Kraus and Woodman’s Greece and Rome survey of the Latin
historians (1997: 6), “it is taken for granted that the views broadly associated with Wiseman and
Woodman are correct” and that “since these ancient texts are as much literary as historical, a
literary approach, in which one reads for structure, style, and theme (among other things), can
offer new insights.” But the controversy is far older: Brunt (1993), whom Kraus and Woodman
identify as their stalking horse, was replying to a previous, mostly German, scholarship (see esp.
his first, unnumbered note) which anticipated some of Wiseman’s and Woodman’s arguments.

41

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 j. e. lendon

did so to such a degree that nearly all of the tradition surviving to us is the
result of successive waves of fabulation.2 If this is true, much of the traditional
practice of Roman history is impossible.3 And so criticism of the Latin
historiography movement by a Roman historian can hardly fail to appear
special pleading: I and mine have a vested interest in the over-all soundness of
the stories about Rome preserved by the Latin historians.4 But a historian – an
outsider – is called to this task because of the lack of self-criticism from within
the Latin historiography movement: it is wonderful how its paladins fail to
apply to each other the astringents and corrosives they bring to bleach Livy
and Tacitus. The outsider is puzzled by this seeming suspension of the normal
process of academic self-correction, and perplexed by prickly classicists evol-
ving a culture of mutual unction. But historians too have not answered as
vigorously as they ought. Unconquerable love of ease is no doubt part of the
reason for this long neglect, but more powerful is the admirable inclination
of most historians simply to get on with it and not worry too much about the
theoretical basis of what they are doing: theirs is the hard-skulled practical
habit of mind that simply ignored Hayden White, and preserved academic
history from the squalls of nonsense from France that overwhelmed the
modern languages. In its first generation, when Wiseman and Woodman
cried out alone, the movement was hardly noticed. Now that Latin historio-
graphy has established itself in their wake as an independent discipline,
ancient historians feel they have the same license to ignore it as they have
to ignore scholarship on Callimachus. Besides, many of the products of this
movement are so removed from conventional history in purpose and lan-
guage – “Is misreading the end of all reading?”5 “This power is power over
meaning: ‘Wor(l)d power’. The misnomer maiestas de-stabilizes Roman dis-
course. Systematically. Into pieces.”6 – as to be an invitation to historians to
stay strictly away.
My contention here is that most contemporary writing in English about the
Latin historians slights those historians’ concern for truth, and that indiffer-
ence to so important a part of what the historians thought they were doing
constitutes a pervasive affliction in the scholarship, infecting the cogency even
of writing on topics – like the historians’ style – to which the truth or falsity of
the historian’s reports is strictly irrelevant.

2
Excluded from consideration here are speeches included in the Roman historians, understood
by ancient readers to be the author’s free compositions.
3
But not, as Damon (2007: 440) points out, social or cultural history, where “plausible is almost
better than true, insofar as it provides evidence about the world-view of the author and his
contemporaries” – which may account for some historians’ indifference to this movement.
4
As Woodman 2001: 338 tartly notes. 5 O’Gorman 2000: 97.
6
Henderson 1998(1990): 275.

42

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 Against Roman historiography

Of course Latin historians frequently failed to tell the truth in fact, for any
of a hundred reasons. Sometimes they failed to tell the truth on purpose.
But for the most part, and as a body, they tried in their narratives to tell the
truth as they understood it. The first part of this essay is, therefore, devoted to
examination of Wiseman’s and Woodman’s arguments to the contrary.
Considered next is the assumption – usually unexpressed – that unifies so
much writing on the Latin historians after Wiseman and Woodman: that in
the absence of truth as a polestar by which to navigate, the Latin historian
wrote with the freedom we normally attribute to creative writers of literature –
writers of novels, of plays, of poems – and therefore that the types of criticism
we think appropriate to those genres, and to ancient rhetoric, can profitably
be applied to the Latin historians. Against that vision I argue that whether or
not the Latin historians in fact achieved truth, the truth-orientation of their
craft controlled or influenced nearly everything they did: that history was in
fact a genre of its own, with its own rules. And therefore to study the Latin
historians as if they wrote “literature” in a modern sense is not only perverse,
but blinds the student to what sets the historians aside from other ancient
writers, to what makes the historians interesting. Finally, to understand the
writing of Roman history not as free creation but as a constrained art – where
the author practiced his creativity within a tight box of acknowledged fact, of
the tradition upon which he drew, and of the audience’s expectations – offers
an escape from some of the more anachronistic and solipsistic of today’s
analyses of the Latin historians.

Well-founded doubts about the tales of old Rome extended back into the
nineteenth century. In Clio’s Cosmetics (1979), T. P. Wiseman proposed a
new mechanism for the production of a highly detailed fictional early narra-
tive. This mechanism was the common education of all Roman historians
of the late Republic and after: rhetoric. Rhetoric taught – and encouraged
admiration for – elaborate, moving, and plausible (but invented) storylines.
A series of rhetorically educated authors of the late second and first centuries
BC, Wiseman argued – the so-called later Annalists, and especially Cn. Gellius
and Valerius Antias – had taken a microscopic body of genuinely old material
and successively elaborated upon it until it reached the enormous volume
represented for us by Livy and Dionysius of Halicarnassus. And we can
witness this process of stuffing – prod the doomed goose and marvel at
the waxing of its liver – by noting how early histories of few volumes were
succeeded by later histories of many without (Wiseman contends) any accre-
tion of real knowledge about early Rome.7

7
For a concise collection of the evidence for this expansion, Oakley 1997–2005: I 72–3.

43

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 j. e. lendon

One historian cared enough to object: Tim Cornell was one of the very few
English speakers who worked on early Rome, and he took up his shillelagh
for the soundness of the tradition.8 To Wiseman’s thesis he responded that we
know so little about the Annalists – whose works survive only in sad gobbets,
and in the use later authors made of them – that it is impossible to evaluate
their methods for good or ill: it is therefore quite arbitrary to name them
fabricators of the bulk of the tradition about early Rome.9 Cornell found it
difficult, moreover, to imagine later Annalists being able radically to diverge
from the versions given by the early Annalists, or the well-known and well-
loved traditions of their city, particularly given the extreme conservatism
of Roman culture.10 At least the very broadest outlines of the literary tradi-
tion about early Rome, he argued, seem to be confirmed by archaeology.11
Rhetorical elaboration – and so the lengthening of accounts – is anyway
possible without doing violence to the basic facts of the story, by adding the
speeches and moral reflections and purple passages which ancient readers
were, and modern readers are, well able to discount.12 A process of rhetorical
fabrication by many hands would have produced many drastically different
stories – but in fact the tradition about early Rome, as we receive it, is quite
consistent. Moreover, when we can see a Latin historian of early Rome at
work in the late first century BC – Livy – he is quite innocent of the large-
scale invention of which Wiseman accuses his predecessors. Livy’s method we
can study very exactly, since we possess great stretches of the parallel text of
one of his sources, Polybius, and nearly all he can be convicted of is a little
innocent rhetorical gussying and cosmetic surgery upon the wartier and
craggier of Rome’s historical titans.13 What license have we, then, to accuse
Livy’s predecessors of systematic fabrication?

8
Cornell 1982 (replied to by Wiseman 1983 (1987)), Cornell 1986a, and, the best formulation,
Cornell 1986b. It should also be noted that the confidence in the tradition Wiseman displays in
his own recent work, and the way he uses the details of that tradition, cannot be reconciled
with the doubt he expressed earlier in his career: see, e.g., Wiseman 1995a: 117, 136; 2004:
37–40, 48 (he believes that the Tarquins were in fact descended from Demaratus of Corinth,
something not even Cornell [1995: 124–5] believes); 58, 64, 66, 71, 72.
9
To which Wiseman might respond that he has argued in detail for the fabrications of
individual annalists: Valerius Antias in 1979: 112–35; 1998a: 75–89; Q. Aelius Tubero in
1979: 135–9; and Licinius Macer in 1995a: 108, 143–4; 2002a: 295–301; 2004: 199–200.
10
Cf. Kraus 1994a: 15, “the existing density of tradition constrains both historian and reader
(we already ‘know’ what has happened)”. Wiseman 1983(1987) responded to Cornell that
loyalty to oral or written tradition does not matter much if that tradition is rubbish.
11
Wiseman would disagree: 1994: 6; 1998a: 76; and by implication 2000 (2008): 267–83; 2004:
87–118.
12
Sen. Suas. 6.14–19 reveals a Roman historian in this process of sifting, on which see Damon
2007: 445–6; she goes on to argue for the ability of modern readers to do the same, 446–50.
13
Luce 1989a: 177; Damon 2007: 444. Nor can Wiseman in fairness protest that Livy did not
need to elaborate because it had already been done by his predecessors: for if rhetorical

44

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 Against Roman historiography

A fair referee would, I think, judge Cornell the victor in this exchange.
But Wiseman was lucky in his opponent. For when Cornell actually applied
his principles – in The Beginnings of Rome (1995) – he sometimes placed in
the literary details about early Rome a faith that would not have been exotic
in the Children’s Crusade. His argument against Wiseman was thus weak-
ened in retrospect. Yet Cornell could have said more against Wiseman.
First, Wiseman’s argument is circular: his denial that we can know much
about early Rome is based on his sense of the nature of early Roman history-
writing and the inadequacy of Roman record-keeping – a sense grounded in
the history of early Rome about which he thinks we have so little reliable
information. Second, other reasons for the “expansion of the past” are
possible, and may be more important than rhetorical elaboration. Even if it
is now fashionable to doubt the old theory that larger histories were made
possible by the publication of the Annales Maximi in the late second century
BC, the mid-first-century Licinius Macer at least seems to have found new
evidence about Rome’s early history: books written on linen in the temple of
Moneta, which Aelius Tubero also exploited in the 30s BC. Actual knowl-
edge of the past was expanded by research.14 And the researches of antiquar-
ians into rituals, inscriptions, and old monuments – researches upon which
we can see Livy draw – would also over time have expanded the amount of
information available to historians.15 The historical value of such matter is,
and is likely to remain, controversial; but such material does provide an
alternative explanation to the theory of rhetorical elaboration for the increase
in the volume of early Roman history.
Third, Wiseman’s theory of rhetorical elaboration for the expansion of
early Roman history may not be fully equal to the task. He suggests – to
impose a metaphor upon him – that an exiguous thread of real material was
dipped repeatedly into a vat of wax by many rhetorical hands – and that the
result was eventually the noble devotional candle familiar to us, capable of
burning steadily for a long time, and elaborately adorned with detail. But
rhetorical magnification is arguably not like candle-dipping. It is more like the
making of pearls, in which nearly all the sea’s sands are left alone, but a tiny
few lucky grains are enoystered, built upon layer after layer, and polished to a
gleaming finish. If the tradition about early Rome had been contrived chiefly
by rhetorical elaboration it would look rather different than it does. There
would be a handful of wonderful stories expanded to enormous length – and

elaboration was admired, it would hardly have stopped in Livy’s day, and he would a fortiori
have applied it to the parts he took from Polybius.
14
For the Annales Maximi, about which controversy will never cease, Cornell 1986a: 71; Frier 1979
(1999). Libri lintei: Livy 4.7.11–12, 4.20.8, 4.23.2; cf. Oakley 1997–2005: I 27–8; Cornell 1995: 7.
15
Cornell 1995: 18–26; Oakley 1997–2005: I 33–7; cf. Wiseman 1979: 45.

45

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 j. e. lendon

little in between. There are indeed stories expanded to great length (and in
such cases we may indeed suspect that rhetoric was at work). But there is all
too much in between. Livy has something to tell us about nearly every year
since the beginning of the Republic. All those election results, minor wars on
minor folk, all that trivia of every type is what makes Livy punitively boring;
but the wretched plainness and repetitiveness of the material suggests that
rhetorical elaboration cannot explain it.16 The end of rhetorical inventio,
Wiseman said, was to “turn the bare annals into a narrative that was morally
exemplary, politically significant, or just dramatic and exciting for an audi-
ence to listen to.”17 Into which of those categories, one wonders, falls the
story that in the consulship of P. Plautius Proculus and P. Cornelius Scapula
(328 BC) a distribution of meat was given to the people by the obscure
Marcus Flavius at the funeral of his mother (Livy 8.22.2)? How about the fact
that in the consulship of L. Valerius Potitus and M. Manlius (392 BC) the
men of Volsinii and the Sapienates – a folk not otherwise known18 – raided
Roman territory (Livy 5.31.5)? Lying between the matter that can plausibly be
ascribed to rhetorical elaboration and the small quantity of documentary
evidence (of whatever reliability) Wiseman is prepared to admit – chiefly
reconstructed lists of magistrates and triumphs – there is too much material
unaccounted for.19 The same yards of faded gabardine also suggest that
patriotic or moral myth-making – as older scholars proposed – was not
responsible in great part for the tradition either: myth-making too would
have raised noble mountains, not dug the endless tumbleweed valleys of, for
example, books vi–x of Livy. How, then, are we to account for this sad stuff,
other than it came down in some sort of written record?
In the 1990s, Wiseman moved on from investigating the rhetorical elabora-
tion of the tradition in the late Republic to consider the nature of the earlier
tradition available for such elaboration. In his Oxford Ronald Syme lecture of
1993 (Wiseman 1994: 1–22), in Remus (1995a: 129–44), in Roman Drama
and Roman History (1998a), and in The Myths of Rome (2004: esp. 147, 248)
he revived an old theory that those early stories were created and transmitted in
the context of Roman plays on Roman subjects given at Rome’s frequent
festivals: “I should like to propose, in place of the ‘history from documents’
idea, which I believe to be untenable, an alternative model of ‘history from

16
For a catalogue of such jejune material in Livy vi–x, Oakley 1997–2005: I 38–72 and esp. 57–62.
17
1981 (1987): 389.
18
But see the speculations of Ogilvie 1965 (1970): 695–6 about location and the correct spelling
of the name.
19
Magistrate lists and triumphs, Wiseman 1979: 13–21; 1981(1987): 389; 1983(1987): “names
of magistrates and triumphatores, a certain number of episodes more or less accurately
remembered” (20).

46

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 Against Roman historiography

dramatic fiction.’”20 As a contribution to some part of the tradition, this is

unobjectionable, even appealing – and it is rather a pity that it has not been
picked up by scholars with the same enthusiasm as Wiseman’s theory of rheto-
rical elaboration.21 Yet to argue that “the Roman historical tradition was largely
created and perpetuated in dramatic performances” (1998a: ix–x; my italics)
runs Wiseman into the same problem as his earlier theory of rhetorical elabora-
tion: for a national history assembled largely from plays is likely to have even
more of a stop-the-presses texture than a national history elaborated for the
most part in rhetoric. If an author were to try to write the history of the Wars of
the Roses from Shakespeare’s history plays, the product would not have the
texture or structure of – or contain the massed trivia that characterizes – Livy.
To the historian, the particular appeal of Wiseman’s elegant theory of
early Roman drama is that it conjures in the mind strong arguments for the
rehabilitation of the old explanation for the early Roman historical record,
exactly that “history from documents” that Wiseman hoped that his “history
from dramatic fiction” might replace. Wiseman finds the context of early
Roman drama in the Hellenized culture of early Rome, and for this mixed
culture his arguments are very strong.22 Very well: but few Greek cities had
traditions of play-writing, while all had civic records. And so the more Greek
influence we see upon early Rome, the more reason we have to suspect that
Rome of the fifth, fourth, and third centuries BC had at least the same
records as contemporary Greek cities, which were in many cases volumi-
nous.23 Nor is it easy to imagine a priori, and it would be historically
unexampled in fact, that a complex, expansionist state like Rome
got along without at least records of public income and expenditure, public
contracts, laws, lists of those liable to military service, and treaties; nor were
the Romans given to tidily throwing old things out; nor, despite many fires,
did the Romans ever suffer a general cataclysm of the sort that might wipe out
such records.24 Wiseman’s portrait of a merrily lascivious, play-going,
Hellenized early Rome reminds us that the idea of archaic Rome as a place

20
Wiseman 1994: 5. But Flower 1995: esp. 173, thinks this was too sparse a tradition to play
much of a role in Roman historical knowledge; Keaveney 2003 and 2006 also attacks
Wiseman’s attribution of specific historical stories to plays, and offers references to other
contributors to the controversy. In support of Wiseman, Kragelund 2002.
21
But it has been picked up, somewhat ironically, by Cornell: 1995: 11–12; 2003.
22
This theory of strong Greek influence going back to Rome’s beginnings has won wide
acceptance; see, e.g., Feeney 2005.
23
Records that could perhaps be augmented, as Cornell 1995: 7–9 and Purcell 2003: 24–6
remind us, by Greek writings that contained information about early Rome.
24
Oakley 1997–2005: I 25: “it would be surprising in a partly hellenized and partly literate society
if the state did not keep records of some kind.” Cornell 1991: 25–32 collects Roman traditions
about official writing in early Rome. Lack of cataclysm: Cornell 1995: 24, 318; Wiseman

47

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 j. e. lendon

without records is no more than a stray wisp of the Romans’ own myth of the
agrarian purity of their past; they even conjured for themselves a vision of an
archaic legal system without documents, another nostalgic fantasy.25
There is not much evidence for the direct use of old documents by the
surviving historians, Livy and Dionysius of Halicarnassus – although there
are traces of such use by the predecessors upon whom Livy and Dionysius
drew.26 But here again Wiseman comes to the rescue: for there is no direct
evidence in those authors for the use of the early drama he posits either. But he
teaches us how such traces might be smoothed away over the generations of
historians using each others’ works which culminated in Livy and Dionysius:
“I suggest that such a play was indeed produced at the Liberalia . . . and that
a generation later a Hellenized historian . . . transcribed it as the detailed
narrative of another episode glorifying his famous family. From there it was
taken over into the Latin historiographical tradition, and thus at last to the
text of Livy” (1998a: 48). We can add that the process of “historical” literary
treatment, as ancient historians understood it, normally included making the
origins of the material they used (be it drama or documents) invisible after the
first generation of use;27 indeed, the first authors to use such material may
themselves have made their own use of plays or documents as invisible to their
readers as possible – and users of documents even more so than the users of
plays. For the high tradition of the genre in which they worked emphasized
dignified researches like talking to individual informants – preferably of high
status – and reading polished literary works; perhaps significantly, the types
of testimonia about early Rome mentioned by Cicero emphasized dignified
2004: 90. But Wiseman does not draw this conclusion about early records from his discussion
of the early Hellenization of Rome. Romans are still “becoming used to the art of writing” in
the 290s (Wiseman 2004: 121) and had no significant records before the 390s (ibid., 90)
because “reliable records are characteristic of constitutionally organized administrations, and
Rome achieved that status only in the fourth century BC” – rather a puzzling remark about the
city able to build the gigantic temple of Jupiter Optimus Maximus in the sixth century. Livy
(6.1.2), like perhaps Claudius Quadrigarius before him (Plut. Numa 1.1), used the Gallic sack
to excuse the lack of records before 390 BC, and Wiseman 2004: 90 argues that this must
mean, even if we now know from archeology that the city did not actually burn, that the
Romans had little written evidence – not even an annual list of magistrates – from before 390.
But the mass of scholarly opinion supports the reliability and antiquity of the pre-390
magistrate lists (Cornell 1995: 13), and Livy has much year-by-year information even before
390 BC: “the nature of the evidence for the period covered by book v [403–390 BC] is not
notably inferior to that for the period covered by book vi [389–67 BC]” (Oakley 1997–2005:
I 39). Whatever Livy thought, 390 BC was not a significant historiographical juncture.
25
Roman myth of agrarian purity, cf. Wiseman 2000 (2008): 287–9; law and documents, Meyer
2004.
26
Collected at Oakley 1997–2005: I 24–72.
27
An historian might comment on his sources, but not usually on his sources’ sources: cf. Walsh
1961: 114: “[t]he normal procedure was to name an authority only in criticism of his account,
or when indicating alternative versions of an event described.”

48

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 Against Roman historiography

sources, except for the Annales Maximi, for which he felt distaste (De Leg. 1.8;
cf. De Orat. 2.51–3).28 About the actual business of assembling history from
very old records – teetering on ladders to peer at dark tablets high on walls,
pleading with public slaves to look again in the mongrel piles at the back of
temples, the dirt, the dead worms – the less said the better; after all, if painting
got Fabius Pictor a rude nickname, one hardly wanted to mention exercises
even humbler.
Both Wiseman’s positive argument for the plausibility of the influence of
early drama on Roman historiography and the subtle way he argues away the
difficulties of evidence it presents can be employed better to fortify the old
theory of the ultimate origins of early Roman history in documents. And since
there is direct evidence for the use of such documents in the tradition (how-
ever scanty), and because there are volumes of material in the tradition –
Marcus Flavius’ distribution of meat and its countless grey brethren – that are
unlikely to go back to plays or rhetorical elaboration, it is probably safest to
conclude that the Roman historical tradition was documentary in its origins,
as Cicero (De Leg. 1.8; De Orat. 2.52) and Livy (6.1.2) thought – however
much contaminated by drama, rhetoric, or any of the manifold traditional
sources of corruption to which scholars have been alert for more than a
century.29

Wiseman confined himself to the easiest case: the history of early Rome. In
Rhetoric in Classical Historiography, published in 1988, A. J. Woodman
ambitiously attacked the credibility of the whole ancient historical tradi-
tion.30 His argument was twofold. First, he maintained that historians in
antiquity did not aim in their accounts for positive truthfulness as we know
it; and second, he argued that the writing of history “was regarded by the
ancients . . . as a branch of rhetoric” (x) – that “classical historiography . . .
is primarily a rhetorical genre and is to be classified (in modern terms) as

28
The much-discussed speculations of Cicero about the nature of the traditions about early
Rome – especially the Annales Maximi, the ballads, and the self-aggrandizing oral traditions of
noble families (see Oakley 1997–2005: I 22–38) – should never, in the process of debating
their value yea or nay, have invisibly been allowed to define the limits of the possible. How on
earth did Cicero know? Cornell (1995: 15–16) is right to stress the existence of records other
than the Annales Maximi.
29
Which does not, of course, mean that the documentary sources were necessarily true, or rightly
interpreted by intermediaries, or that we can easily detect contamination: extreme care is
necessary in accepting the traditions about early Rome. I would affiliate myself with the
reasoned caution of Oakley 1997–2005: I 3–110; cf. Kraus 1994a: 28.
30
Including the Greeks, and especially Thucydides (1–69), usually thought the most accurate of
them, a line of argument I ignore here except to point out that it is answered by Brock’s review
(1991).

49

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 j. e. lendon

literature rather than history” (197). Rhetorical techniques – especially rheto-

rical inventio – were allowed free play by ancient historians, indeed encouraged
by the rules of the genre. Fabrication was allowed in theory and frequent in
practice. On the theoretical level Woodman’s argument depends on his inge-
nious readings of seemingly straightforward ancient discussions of historical
method, in the Roman case those of Cicero. The orator offers an implicit
definition of the “laws of history” and historical truth (Ad Fam. 5.12.3) in a
letter urging a friend to set them aside – when praising Cicero – and his
character M. Antonius returns to the subject in the De Oratore:31 “Who
doesn’t know that the first law of history is not to dare to say anything false,
and the second is not to refrain from saying anything true?” (2.62).32
To Woodman, Cicero’s immediate sequel to that statement, “there should
be no prejudice for, or bias against, when you write” is an exhaustive gloss of
the Roman historian’s conception of true and false: avoiding bias is all there
is to it.33 But John Marincola has gracefully corrected him. Ancient historians
were indeed anxious about bias, but that was because bias led to making
things up, led to divergence from what really happened.34 Roman historians’
understanding of truth was not in fact impoverished: it had both the negative
(avoid bias) and positive (say what really happened) content of the wider
ancient conception of truth – that possessed by people who were not histor-
ians – and of our familiar, modern, man-in-the-street conception of truth.35
Cicero’s character was merely providing the most likely motivations for a
historian’s straying from the truth, and Woodman is reading a definition
where none was intended – while failing to find an included concept that
only an unnaturally beady-eyed reader would ever imagine needed to be
specified. “Jim got in his car, turned the headlights on, and drove away.”
Woodman: “the author’s failure to specify that Jim turned on his engine
shows us that the twentieth-century concept of the automobile included
only headlights, and not the engine.”36

31
For a more straightforward reading of De Orat. 2.51–64, with which I would concur, Leeman,
Pinkster, and Nelson 1985: 249–52.
32
Cf. Cic. Brutus 42, “concessum est rhetoribus ementiri in historiis,” followed by criticism of
fabulating historians: so it was not conceded to historians to lie (cf. Brunt 1993: 201; Potter
1999: 137).
33
“ne qua suspicio gratiae sit in scribendo, ne qua simultatis” (trans. Woodman). Cf. Wiseman
1981(1987): 387–8 and Wiseman 1993: 126–7.
34
On ancient understandings of historical bias see also Luce 1989b.
35
Marincola 1997: 160–2; Marincola (2007a); Marincola (this volume): 18–19; Potter 1999:
12–18; Morgan 1992–3: 35.
36
Blockley 2001 shows that Ammianus’ conception of historical truth was not so impoverished
(as Woodman 1988: 101–2, n. 7 himself hinted), and argues that Woodman has misunder-
stood Cicero.

50

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 Against Roman historiography

Woodman goes on to argue that Cicero’s subsequent description (still placed

in the mouth of Antonius) of how an orator should write history shows that
history was accepted as a branch of rhetoric and so subject to the blameless
falsification that characterized ancient speech-making, both in theory and in
practice.37 He is quite right that Cicero’s description of what constitutes the
res (the matter) of history sounds very like his descriptions of rhetorical
narratio in which plausibility was the goal and inventio was given full rein
(Inv. 1.29; Part. Or. 31–2). But Woodman’s rigorous subordination of the
first and second laws of history – say nothing false; say all that is true, and
“these foundations are of course recognized by anyone” (2.63) – to this
discussion of res has been broadly questioned, and rightly.38 And the wider
context of the passage defeats the conclusions Woodman would draw from it.
History in Latin is, alas, not currently (Cicero has Antonius say) a branch of
oratory.39 Would that it were! And if it were, here is how an orator should
go about writing it.40 Antonius’ prescriptions for how to write history in
Latin are a counterfactual speculation, and orators writing history would be a
new departure in the world.41 For Antonius notes that the Greek historians,
although often eloquent, were not in the Latin sense orators either: “among
the Greeks the most eloquent men had nothing to do with speaking in the
courts, but applied themselves to other noble things, and especially to writing
history” (2.55). He then offers a list of Greek historians – Herodotus,

37
Woodman re-states part of this argument, briefly and rather more clearly, in Woodman 2001:
342–3.
38
Leeman 1989: 238; Morgan 1992–3: 35–6; Blockley 2001: 22; Bosworth 2003: 169. See esp.
Cic. De Leg. 1.5 (with Dyke 2004: 74), where (however one resolves the textual issue), Cicero
seems clearly to indicate that truth is fundamental to history and everything else secondary:
“alias in historia leges observandas putare, alias in poemate.” “Quippe, cum in illa <omnia>
ad veritatem, Quinte, referantur, in hoc ad delectationem pleraque”; also De Orat. 2.36:
“historia vero testis temporum, lux veritatis . . .”; 2.51: “non esse mendacem.”
39
Alternatively, as Wisse 2002a: 361 reads it, “Antonius does say that history is a great task for
the orator (2.62, cf. 51), but that by no means implies that it is a task for rhetorical theory. On
the contrary, his point is that for an orator who is a master of the judicial genre, which he
considers the most difficult, no further instruction is necessary, whatever task he may take up,
such as history. Therefore, Antonius is arguing, it is only natural that rhetorical theory does
not cater for such other tasks.” Whether Antonius regrets (Wiseman 1979: 31; Woodman
1988: 104–5) that rhetorical theory does not comprehend history, or does not think it needs to
(Wisse), the point is that it does not. And if Antonius thinks that it should, even that can hardly
be taken as a generally accepted view: Wisse 2002b shows how self-consciously anti-
conventional the views expressed in the De Orat. are.
40
Leeman 1989 argues that the discussion of writing history is only included at all, and takes the
form it does, because Cicero is depicting an imaginary perfect orator (1.118), who combines in
his person mastery of all forms of expression (2.45, 49–50).
41
Cf. Feldherr 2003: 198, “the dialogue presents the idea that history should be written by an
orator as something radically new and fundamentally different from the way history has been
produced in Rome.” Also, Cape 1997: 219.

51

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 j. e. lendon

Thucydides, Philistus, Theopompus, Ephorus, Timaeus – to show that none

of them pled in the courts, that none of them were orators (2.55–8), even
though Ephorus and Theopompus were trained by the orator Isocrates (2.57)
and Callisthenes wrote in an “almost rhetorical style” (2.58, “rhetorico paene
more”).42 Nor, Antonius says, could one find instructions for writing history
in rhetorical handbooks (2.62, 64) – nearly all in Greek at both the dramatic
date of the dialogue, 91 BC, and the date the dialogue was written, 55 BC
(Quint. Inst. 3.1.19–20; Cic. De Orat. 1.23).43 In classifying history as a
subset of oratory Cicero was not describing something real in his day either in
the Greek or Roman worlds (he is quite emphatic on that point), but instead
something he wished might be the case – offering a persuasive, rather than
factual, classification; hardly a surprising piece of intellectual imperialism for
the leading orator of his day, upon whom (he tells us elsewhere) writing
history had been urged (De Leg. 1.5).44
Nor was Cicero the only such rhetorical imperialist. In his La storiographia
nell’educazione antica (1992) Roberto Nicolai placed Cicero in the context of
a swirling debate among rhetorical theorists – chiefly Greek – about the place
of history in their intellectual system. Was it to be classified as deliberative
oratory? Epideictic oratory (cf. Cic. Orat. 37, 66)? Was it to be granted a
category of its own?45 Which historians were to be categorized as high style,
medium style, and low style? Oh, the exquisite distinctions, the ecstasy of
parsing! But silent in this wrangle are the historical authors themselves. They
wanted no part in rhetorical theory’s attempt to gobble up the whole intellec-
tual world like a rubber monster ravaging a plastic Tokyo.

42
Antonius’ list is not only “a potted history of Greek historiography in order to contrast its
rhetorical nature with the unrhetorical nature of Roman historiography” (Woodman 1988:
78), but primarily an account to show that Greek historiography is eloquent without being
written by orators as the Romans understood them, although, Antonius argues, if the Romans
want to match the Greek achievement, it will take a great Roman orator to do so (2.51, 2.62).
Woodman’s interpretation of this passage cannot account for Antonius’ repeated stress on
how different Greek historians are from Latin orators.
43
For history not being regarded as a branch of rhetoric in the Greek tradition, Oakley 1997–
2005: I 7; Marincola 1997: 161–2; Wisse 2002a: 361. But contra, Nicolai 1992: 89–155.
44
And Cicero might find history appealing to take up because “sit opus, ut tibi quidem videri
solet, unum hoc oratorium maxime” (1.5). But again, this is a contrary to fact: Cicero has not
in fact taken up writing history (and the deficiencies of Roman writers past and present are
soon canvassed), so the project of unifying the two arts is merely a pipe-dream, not something
that has happened in fact. Against Cicero regarding history as a branch of oratory cf. Wiseman
1979: 31–4, citing Cic. Brutus 286: “non tam historico quam oratorio genere”; Brunt 1993:
197, who also argues more broadly against older scholarly attempts to classify history as part
of rhetoric, and Potter’s (1999: 135–8) polemic against Woodman, but that should be read
with Woodman’s magnificent review of Potter (Woodman 1998a), and Oakley’s 1997–2005:
I 7–9 defense of parts of Woodman’s position.
45
Nicolai 1992: 89–155.

52

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 Against Roman historiography

Yet was Cicero prophetic? Even if the genres of history and oratory had
not become confused by Cicero’s day, had they done so by the ages of Livy or
Tacitus? Rhetoric was the universal education of the classes from which histor-
ians were drawn: the influence of rhetoric on historians’ style can hardly be over-
stated. But does that mean that the two arts ever became indistinguishable to
Romans? Not according to Quintilian (c. AD 96), who cautiously recommends
to the young orator the reading of history, but bids him beware: because history
and oratory are so different in purpose and style (Inst. 10.1.31–4; cf. 2.4.2), the
orator must take care similar to that used when reading poetry and philosophy
(Inst. 10.1.27–30, 35–6). Like Cicero, Quintilian evaluates historians in part by
the similarity of their style to oratory (Theopompus: “oratori magis similis”;
Inst. 10.1.74) which indicates that to him the genre boundaries between them
were still perfectly visible. They were equally visible to his contemporary Pliny
the Younger (Ep. 5.8). Those who wrote history in classical antiquity were
perfectly well aware that they were doing something different (which observed
different rules) from writing speeches, plays, poems, or works of philosophy.
So the theoretical distinction between history and oratory was preserved,
and the ancient concept of historical truth (to which, in principle, the histor-
ian was to cleave) was not significantly different from our own. But none of
this is very important if Latin historians simply ignored the theory of their art,
and did in fact simply behave like rhetoricians, inventing and embroidering at
whim. Yet, again, in the sole case where we can clearly see a Latin historian
using his sources – Livy using Polybius – this does not happen.46 And if
historians expected each other to invent, that would render rather perplexing
the tradition, which Latin historians inherited from their Greek predecessors,
of historians praising each other for their truthfulness and reproaching each
other for their departures from truth. Latin historians do this not only in their
prefaces but also in the course of their narratives, correcting their predeces-
sors’ errors en passant, and halting for a moment to vaunt over them.47 In his
surviving books, Livy explicitly or implicitly corrects his predecessor Valerius
Antias on more than twenty occasions.48 And it was not only the historians
who evaluated each other on their truthfulness: rhetoricians evaluated

46
See above n. 13. Bosworth 2003: 175–86 also finds the first-century Quintus Curtius not
guilty, by comparing him to parallel accounts, rather than to his sources (which are lost).
Morgan 1992: 14–26 and Damon 2007: 444–5 also impeach an instance of inventio in Tacitus
alleged by Woodman (1988: 168–79), the description of Germanicus coming upon the scene
of the Varian clades (Ann. 1.61–2).
47
Luce 1989b: 23–7; Marincola 1997: 247–53; Marincola (forthcoming). For praise, esp. Sall.
Hist. 1.4m: “Sallustius . . . in libro primo Historiarum dat . . . Fannio . . . veritatem”; Tac. Ann.
4.34 (speech of Cremutius Cordus): “Titus Livius, eloquentiae ac fidei praeclarus in primis.”
48
Gathered by Oakley 1997–2005: I 89–91. For Tacitus’ correction of the authors upon whom
he drew, Mendell 1957: 199–214.

53

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 j. e. lendon

historians in the same way.49 Cicero praises Polybius for his accuracy (Rep.
2.27; Off. 3.113) and denounces other Greek historians for their mendacity
(Brutus 42–3; De Leg. 1.5).50 For the most part, naturally, Quintilian eval-
uates historians for their style, but even Quintilian notes that “the genius of
Cleitarchus is praised, but his trustworthiness is abused” (“probatur inge-
nium, fides infamatur”; Inst. 10.1.75).
Historians not only commented on each other’s veracity, implying its
importance to them, but far more important, they used each other as true:
they drew upon the work of predecessors critically, but for the most part
without pervasive doubt about the accuracy of those predecessors.51 If con-
fidence in the truth of the tradition was for a systemic reason (such as the
influence of rhetoric) unjustified, who would have known it better than other
historians, who participated in that same tradition? But those in a position to
know best had confidence in each other.
Obliteration in practice of the distinction between truth-oriented history
and plausibility-oriented rhetoric would also render rather mysterious Latin
historians’ habit – also inherited from the Greeks – of visibly pondering which
of two contradictory versions of a story to believe.52 “Licinius Macer and
Tubero report . . . but I do not put it down for certain because Piso, an older
author of annals, declares . . .” (Livy 10.9.10–12; cf. Tac. Ann. 14.2; Hist.
3.28). We can scream with mirth at the feebleness of the criteria sometimes
used to make such decisions – we often want to throw out both versions53 –
but the very fact that historians puzzled over the relative value of their sources
shows that they were concerned to produce a truthful account. The standards
of rhetoric, had they dominated history, would have encouraged no such
deliberation. Nor would they have encouraged expressions of dissatisfaction
with a previous author’s account, even in the absence of an alternative
account to follow – “Since I have no other authority for this rumor I can
neither confirm it nor deny it” (Livy 37.48.7; cf. Tac. Hist. 2.37)54 – nor
admissions of bafflement – “I will follow the authors when they agree, and
where they provide different accounts, I will record them under their names”
(Tac. Ann. 13.20) – or the sense, which one gets so frequently in all the Latin
historians, of an author straining to put together an account in the face of

49
For ancient criticism of the mendacity of historians, Wiseman 1993 collects the references. The
habit of criticism on this point implies, of course, that historians were expected to tell the truth.
50
For Cicero’s evaluations of historians, Fleck 1993. 51 Bosworth 2003: 170–2.
52
In Livy vi–x, Oakley 1997–2005: I 13–15; also Sen. Suas. 6.14–19 with Damon 2007: 446.
53
Wiseman 1979: 48–51.
54
Seneca Quaest. Nat. 4.3.1 mocks this habit as a mendacious way of conveying verisimilitude
upon the rest of the text (see Wiseman 1993: 135); but it only works as such if readers accepted
that such expressions were normally sincere.

54

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 Against Roman historiography

intractable or inadequate evidence. If the louche rules of rhetoric had been

all historians obeyed, they would not have had to work so hard.
How was the overall truthfulness of history writing patrolled? Those
writing the history of their own times were constrained, of course, by the
scoffing memories of their contemporaries.55 Those writing of generations
before their own were constrained, in part, by the competitiveness that
history-writing shared with all other ancient forms of upper-class intellectual
endeavor. An author felt himself competing with “a great mob of writers”
(Livy 1. praef. 3), and historians competed both in style and in getting the
facts right: “the succession of new writers always thinks that they can
provide more certainty in their facts or surpass rude antiquity in their style
of writing” (Livy 1. praef. 2).56 And that some process of regulation – whether
by competition between authors, self-regulation by authors, discipline by
fear of a broader public opinion, or all together – operated is argued by the
self-assured narrative manner of the Latin historians. Historians took it for
granted that their readers would accept nearly everything they said as true
without needing to argue it. The elementary business of the Latin (or Greek,
after Herodotus) historian is to relate a series of events without telling his
reader where he got the information about them, or stopping to say why his
reader should believe him: a convention that only works if the reader already
accepts that the truthfulness of what he is being told is pre-emptively vouched
for by the genre of the work he is reading.57 Otherwise, every statement the
historian made would have to be supported by evidence and eyewitness
testimony, or argued for – and so Latin history would read far more like
Cicero’s Verrines, or Herodotus (who had not the authority of the genre to
fall back upon, because he was creating it). We know what an ancient
historian sounds like – how he reasons, argues, and wheedles – when he is
afraid his readers will not believe him: Thucydides in his Archeology. But, in
fact, “history is written for telling, not proving” as Quintilian put it (Inst.
10.1.31), and the lack of proving in Latin historical works shows that author
and reader knew perfectly well that they were engaged in something quite
different from forensic rhetoric, that history was an undertaking where both
parties took the essential veracity of what was being related for granted.58 It
is all very well to study how readers might be manipulated by historians’
various claims to authority.59 But readers were not infinitely malleable. Had
Latin historians strayed systematically from their readers’ expectation that

55
Damon 2007: 442–3. 56 For competition, cf. Sall. Hist. 1.3m, Justin, Pref. 1.
57
Wheeldon 1989: esp. 44–5.
58
Ancient critics in general are far more interested in ancient historians’ style than in their
veracity, exactly because they took that veracity for granted: Wheeldon 1989: 60–1.
59
See esp. Marincola 1997.

55

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 j. e. lendon

they were being told the truth, had they not vindicated that expectation
generation after generation, the genre would soon have lost the confidence
of its readers, whatever the historians’ claims to authority, and the genre
would have been obliged to convert to a more forensic, arguing, proving
style – or simply to dissolve. Ultimately, the confident, unforensic, narrative
manner of the Latin historians is a strong argument that they thought what
they were saying was true – unless we are prepared to believe that they
successfully perpetrated a centuries-long conspiracy against the public.60
So truth-telling was important to ancient historians not only in theory but
also in practice: and, of course, the fact that most of the narrative after 100
BC preserved in Latin historians is true has never been seriously doubted, nor
can it be: for when historical accounts are compared with other evidence (the
speeches and letters of Cicero for the late Republic, and inscriptions particu-
larly under the empire) they fit together too well to allow the conceit that
Latin history writing was something akin to modern novel writing.61

The claims of Wiseman and Woodman were scanty dragon’s teeth from
which to raise up an army of followers. But if their arguments were not
always sound, they were certainly liberating: and folk believed their argu-
ments regardless of their details because they yearned for them to be true.62
Wiseman and Woodman freed the study of ancient historical texts from the
mortification of being a discipline ancillary to history, like the study of ancient
inscriptions. No longer would the humble historiographer supply Hamburger
Helper at the historians’ barbecue. The Latin historians were made available
to “real” classicists and their literary methods; Wiseman and Woodman’s
conclusions fitted too with Hayden White’s fashionable reduction of history
to a function of literature. Two of the strongest forces, a longing for freedom
and a new-found pride (for the study of Latin prose authors was traditionally
regarded as the province of dullards), drove scholars towards Wiseman and

60
Systematic fraud in a genre close to history-writing did exist: the Historia Augusta, which
purports to be the work of several authors when it is in fact by but one, and where the later
imperial lives are mostly invention. But such an ingenious hoax loses its point if the genre as a
whole did not have credibility: no point to corrupting professional wrestling when everybody
knows that it is already fixed.
61
For a sense of the limited traction Tacitus offers to critics of his general veracity, see the
aggressive comparisons of his versions of the same events to bronze copies of a speech of
Claudius, the Tabula Siarensis, and the Senatus consultum de Cn. Pisone patre (Kraus and
Woodman 1997: 99–102; Woodman 2004: xv–xvi) – rather like taking on a battleship with a
popgun. Nor, if pushed to it, would Woodman deny that nearly all of what Tacitus says is in
fact true: Woodman 2007: 144.
62
“Woodman’s main guideline for distinguishing ancient from modern historiography – admis-
sibility of invented material – has liberated the study of ancient historiography from its
previous positivistic incarnations . . .” (Haynes 2003: 30).

56

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 Against Roman historiography

Woodman; and careerism too soon entered in, as young classicists saw in the
literary study of historians a road to lofty elevation.
What the followers of Wiseman and Woodman have in common is a
stepping aside from the truth of the texts. Some make ritual disavowals of
concern for assessing the truth or falsity of the statements made by the
historians they study; such abjurations are, like frog-green blazers at the
better sort of country club, claims to membership in the school of Wiseman
and Woodman.63 Others affirm that the authors they study were trying to tell
the truth,64 but it does not matter much to these students in practice, for they
have accepted the invitations of Wiseman and Woodman to investigate the
particularly literary quality of historical texts.
Yet this agnosticism about or indifference to truth and falsity in the histor-
ians is the basic weakness of this second generation and their work. For even if
their subject is the historian’s artistry, the constraints the historian believed
applied to that artistry make all the difference in the world: the painter of
imaginary landscapes and the painter of portraits from life may both be artists,
and use some of the same techniques, but no one would think that the differ-
ence in their relationship to real appearances was irrelevant to understanding
their work. Although they may claim agnosticism, or even nod towards the
historians’ concern for truth, the followers of Wiseman and Woodman tend to
analyze as if the historians they study were in fact authors of imaginative
fiction, free to write absolutely as they pleased. But if they were not, if they
were constrained in their depictions either by what actually happened, or by the
tradition of authors before them (whatever their relation to what actually
happened), then what the followers are doing is as perverse as it would be to
read the New York Times as if it were a novel by John Grisham. Scholars are
trapped unaware by the modern binary categories of literature and non-
literature, of fiction and non-fiction, just as the ancient Greek could understand
“slave” and “free” but scratched his head over the frequent empirical existence
of persons whose status was “between slave and free.” We have no useful
category for the realm inhabited by ancient historical texts: rather than being
“literature,” the works of ancient historians came far closer to the modern
genres of the “non-fiction novel” or popular, non-academic history, where a
degree of embroidery and imagination is layered upon a basis of fact.
A second problem, granting for a moment that Latin history can be regarded
as a literary genre, is: which literary genre? “I assume that the AUC [Livy’s

63
Ash 1999: viii; Feldherr 1998: ix–x; Kraus 1994a: 9 n. 39; Marincola 1997: 1.
64
There is a range of opinion (even within the work of a single scholar) on the degree of the
historians’ concern for truth in their narratives: Marincola (this volume): 18–19, Oakley
1997–2005: I 3–108, Levene 1997: xi, and Kraus 1994a: 28 all endorse, in theory, to greater
or lesser degrees, the truth-orientation of the tradition.

57

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 j. e. lendon

Ab Urbe Condita] has a rhetorical basis and didactic aims,” writes Christina
Kraus, so accepting Woodman’s claim that history was a branch of rheto-
ric.65 But if it was not so regarded (and I submit that the evidence is that it was
not), Professor Kraus and those who think like her have, by making a mistake
of genre, missed the opportunity to investigate the nature of the genre in
which a historian did think he was writing. In a world suffused with rhetoric,
the ancient historians were, perhaps, trying to do something different and
interesting: and that special something is ignored in the conviction that they
were instead trying to do something boring and conventional. There is a
flattening, homogenizing, bulldozing-the-eccentric-hills-of-reality strain to
all this – rather as in 1970s Marxist history, in which, whatever the question,
the answer was always “class struggle”; or 1990s Foucault-influenced his-
tory, where the answer was always “power.” It seems rather a pity to cast the
genre one studies, with all its interesting points and prickles, into the smelters
of another genre to be melted down. Suspicion too should attach to any
intellectual movement that allows academics lazily to apply to a new set of
texts the techniques – in this case close reading of oratory and poetry – they
learned in graduate school. It is all too easy to see everywhere what one was
trained to see somewhere – like the becalmed fig-merchant, who looks out in
vexation at the glassy sea, and proclaims “it wants figs!”
Besides the true intellectual offspring of Wiseman and Woodman, there are
also those who use Woodman and Wiseman as a flag of convenience. They
adopt their assumptions in order to make the text of the Latin historian
“literature” in a special sense, not only free of the ballast of facts but free of
the constraints of ancient genre, indeed free of the Roman world – and of
gravity – entirely, and so available for various forms of anachronistic criti-
cism.66 Wiseman and Woodman, as proper classicists, were acutely alert to
anachronism: they called for one ancient genre, history, to be colonized by
other ancient genres, rhetoric and drama. So the adoption of their theories by
this cynical crew must inspire in Wiseman and Woodman the same strange
mixture of horror and pride that a father might feel upon learning that his
fourteen-year-old son has got a classmate with child.
Steven Rutledge (2000) uses postcolonial theory to interpret Tacitus’
Agricola – although his unsurprising conclusion, that Tacitus was partial to
Roman aggression, hardly required such apparatus. To Holly Haynes (2003),
Tacitus in the Histories was striving to provide an analysis of the operation of

65
Kraus 1994a: 9 n. 39.
66
Batstone (this volume), a vitrine of strange postmodern objects of virtù – “do events exist?”;
“We need to jettison ‘causality’”; “Is there closure in the world?” – defends such knowing
anachronism (28–29) on the grounds of the unconscious anachronism of previous generations
of scholars. All one can do is weep.

58

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 Against Roman historiography

political ideology similar to that of the Marxist theorists Žižek and Althusser
(but to get him there requires a detour through the method of Leo Strauss).
To Katherine Clarke (2002), Tacitus’ Annals are Anti-History, posing the
question, “How can we write a history of anything?” But to Ellen O’Gorman
(2000) the Annals are really late-deconstruction musings on the theoretical
difficulties of reading itself. To John Henderson the Annals were – well, as
usual with John Henderson, who can tell? But look at the end of an article
of his (fig.1):67

Yes, here Roman veins spray their page of imperial history with an ‘Absolutely
Free’ Cynic matinée, a sardonic splash for Iuppiter Liberator, showpiece for
posterity. Specta, iuvenis (16.35.3, ‘Catch this, young man’). Make it work!
It will! Here is the ultimate simplification, the finale, ‘exit’, finis of Annalysis,
wor(l)d without***:
post lentitudine exitus grauis cruciatus adferente, obuersis in Demetrium * * *
as the slashes in both arms work s-l-o-w-l-y,
crushingpaincomesonagainandagainpainafterpaincrushing . . .

m i

-t o e e r .

D t . u . . s

. w .

. e

t .

. .

. . u . . .

. .

.r
Figure 1

67
Henderson 1998(1990); quotation on p. 299.

59

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 j. e. lendon

Even leaving the strange wisdom of Henderson aside, is it not a wonder that
Tacitus, an ancient Roman, shared so many of the concerns of late twentieth-
century literary critics? Glorious the man who anticipated Žižek! And it is, I
suppose, fruitless to inquire whether these alleged views can be paralleled in
other ancient authors, or whether ancient people even had the intellectual
armament to think in such terms: the readings of Tacitus required to sub-
stantiate these positions are so arbitrary that similar ones could be applied,
mutatis mutandis, to make any ancient author say nearly anything. The
famous preface to the Histories “implies that the notion that it is possible to
distinguish ideology from nonideology is itself an ideological gesture, because
it involves the intervention of a fiction made-up precisely in order to sustain
belief in its own order.”68 Huh? In the Agricola Tacitus describes Britain as
“Full of nothing.”69 Really?
Writing of this sort, piling absurdity upon anachronism, does not merit
refutation, only inoculation. And inoculation against it begins by grasping
that these are extreme instances of regarding the Latin historian as an artist
who produces “literature” without the check of fact or tradition. Inoculation
continues by restoring in the mind that essential curb, by imagining Tacitus, a
Roman wielding a Roman pen in a Roman room, poring over and comparing
the works of his Roman predecessors, here re-writing, here combining, here
adding something he has found out, here thrilling when he can finish a story
he got from a predecessor with one of the brilliant sententiae his Roman
audience loved. The best way to banish wild flights about what nested in the
minds of Roman historians is to remember how derivative they were – and
that they were derivative because that was the best way they knew of telling
the truth.

What is lost when you remove Roman history from the Roman historians? Lost
first is a sense of what the historians themselves thought they were doing,
narrating events that really happened in the past. However narrowly philo-
logical the ends of such investigation, therefore – however much the scholar’s
interests be limited to studying the historians’ word-choice or sentence struc-
ture or patterns of alliteration – the endeavor must fail, for it wrongly assumes
that an ancient historian had broad freedom in what he was going to say and
in how he was going to say it. Of all the ancient genres, history arguably gave
its practitioner the least freedom to stray from what had gone before – be that
what he and his generation had known and suffered themselves, or the
writings of his predecessors. See how carefully Livy follows Polybius, how
his Latin sentences again and again echo the Greek! If we assume that Tacitus

68 69
Haynes 2003: 37. Rutledge 2000: 76.

60

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
 Against Roman historiography

worked in the same way (and we must), we cannot read him as an uncon-
strained literary artist. The art of the ancient historian was like that of the
sonnet-forger: all about working within constraints. To ignore the most
important of those constraints makes a nonsense out of reading the historians
as literature.
With the excision of history from the historians is lost, second, the cultural
context of the ancient writings: an awareness of how Romans acted and
thought, of what was important to them and what footling, of what ideas
they held and what ideas they could not possibly have held. With this disap-
pears the ability to judge which postulates about Romans (including Roman
authors) might be true, and which postulates about Romans cannot be. Lost is
the ability to distinguish ancient genres, such as history and oratory, because
gone is the power to discern the interests of specific historical actors – rhetorical
theorists – who were motivated to minimize the differences between genres.
And lost is a scholar’s most important external tool for evaluating whether a
given claim about a text is actually true, as scholarship turns away from Rome’s
rocky reality onto the undemanding plain of textuality.
Finally, when history is cast out of the Latin historians, discarded also are
the robust intellectual habits of the modern historian, to be replaced, if the
restraint of stern philology fails, with the weak and whimsical instruments of
the contemporary literary critic. A sense of argument, of proof, of scale, of
proportion, even of logic and coherent language – all depart. Scholarship
becomes indistinguishable from its parody, and the subject of inquiry shifts
from the geysering fascination of antiquity to the dull, trend-obsessed, and
self-obsessed mind of the critic. The result is like the diary of a fat teenager:
riveting only to its creator, repellent to others, and illuminating to none.

Further reading
Most studies of the Roman historians written before the current fashion
for regarding them as authors of imaginative literature – most of what was
published before 1990 – are useful to contemporary historians of Rome.
Without prejudice to many other works, one can point to Mendell 1957 on
Tacitus, Walsh 1961 on Livy, and Syme 1964 on Sallust. Damon 2007 shares
many of my doubts about current trends. The best recent work on the
historians takes the form of commentary where, whatever the author’s posi-
tion on Roman historians as literature, the attention the genre requires to
historical minutiae limits the damage: Oakley 1997–2005.

61

https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010
https://doi.org/10.1017/CCOL9780521854535.004
CambridgePublished online
Collections by Cambridge
Online University
© Cambridge Press
University Press, 2010