Cicero's letter to his brother Quintus from February 54 is best known for containing the sole explicit contemporary reference to Lucretius’ De rerum natura, but it is also notable as the source of the only extant reference of any kind to another (presumably) philosophical didactic poem, Sallustius’ Empedoclea (Q. fr. 2.10(9).3= SB 14):

In addition to the vexed but separate question as to whether the Sallustius in question is to be identified with the historian, with Cicero's friend Cn. Sallustius, or some other figure bearing that nomen, the meaning of the barbed comment on his poem has been almost as fiercely debated.Footnote ¹ The antithesis between uir and homo has been thought problematic, a difficulty formulated with characteristic brusqueness by Housman:

Housman's proposal of a lacuna following uirum te putabo, where a different protasis corresponding to that apodosis has dropped out, earned a place in Shackleton Bailey's apparatus and a ‘fort. rect.’ in Watt's, but has otherwise found little favour.Footnote ³ Most critics have been more or less satisfied that the strict illogicality should not stand in the way of the joke, though several share Housman's related feeling that homo would stand in more natural antithesis with god or beast.Footnote ⁴

It is worth stressing that Housman is, on the question of Latinity at least, quite right that one cannot be a uir if one is not a homo (though the reverse is of course quite possible). Even the vast resources provided by concordances, the TLL, and now searchable electronic databases such as the PHI CD-Rom or the on-line Bibliotheca Teubneriana Latina merely corroborate the accuracy of his Latinity.Footnote ⁵ The juxtaposition of uir and homo is indeed a common one, and particularly so in Cicero. In many instances the same person is (usually) praised using both nouns, each qualified with an adjective which in some cases may partially reflect the distinction between qualities appropriate to a Roman male and the more humane attributes of a Mensch (e.g. hominem honestissimum, uirum fortissimum, Font. 41; forti uiro et sapienti homini, Leg. Man. 20), but in others (the majority) the contrast is often so hard to draw that the words feel almost like synonymous doublets (e.g. consulari homini clarissimo uiro, Verr. 3.184).

When the two words are set in antithesis, it is always clear, and indeed the point of the antithesis or a fortiori argument generally depends on the fact, that to be a homo is a lesser attainment than to be a uir. Thus the gold ring which Verres gave to a scriba proved not that the latter was a brave man, but merely that he was a rich fellow (neque … uirum fortem, sed hominem locupletem esse declarat, Verr. 3.187), the diminution of a proconsul's province should be guarded against not only in the case of a man of the highest standing, but even in that of a middling fellow (neque solum summo in uiro, sed etiam mediocri in homine <ne> accidat prouidendum, Prov. cons. 38), and Lucius’ and Patron's proto-Hobbesian philosophy describes not a good man but a cunning fellow (se de callido homine loqui, non de bono uiro. Att. 7.2.4 = SB 125). Taking the opposite trajectory, from mere homo up to uir, Cicero often self-consciously corrects himself, promoting his subject from the former to the latter category, as with Cato at Brut. 293 (magnum mercule hominem uel potius summum et singularem uirum) or Epicurus at Tusc. 2.44 (homo minime malus uel potius uir optimus).Footnote ⁶ From this it is at least implicit that to be a homo is a necessary but not sufficient condition for being a uir, but that uiri are a subset of homines is absolutely clear when Cicero writes of injustices which would seem intolerable not only to a good man but more broadly to a free human being (ut non modo uiro bono, uerum omnino homini libero uideatur non fuisse toleranda. Inv. rhet. 2.84).Footnote ⁷ Perhaps the closest Cicero comes to a clear distinction is in his consolatio to the exiled Sittius, where he urges him to remember that he is both things (et hominem te et uirum esse, Fam. 5.17.3 = SB 23), a homo because he is subject to the vicissitudes of all humanity, a uir because he ought to bear those vicissitudes with fortitude. Here there is no fusion or explicit overlapping of the categories; each has its specific and discrete associations. However, neither is there anything here to contradict the evidence of all the other instances or to suggest that even Sittius could be a uir but not a homo. Even with the benefit of searchable databases, it can be seen that Housman's judgement on Latinity and logic is sound. It may be, however, that the confounding of logic (and perhaps of Latinity) is the essence of humour, and so we must ask ourselves whether Cicero's transmitted judgement on Sallustius, since it isn't quite Latin, is actually funny.

Even those who defend the paradosis seem vaguely apologetic about the joke which they are determined to preserve. Shackleton Bailey, in refuting Housman, writes that ‘Cicero says these two things in the same breath … because he thought it mildly amusing’, and in his shorter commentary remarks, almost shame-facedly, that ‘the juxtaposition is mildly funny’.Footnote ⁸ Of course, whether the reason lies in cultural contingency or in transhistorical unfunniness, no one who has read any quantity of Ciceronian ‘jokes’ would consider a failure to provoke uproarious laughter as grounds for emendation.Footnote ⁹ Yet the problem with this joke is not so much that it is at best ‘mildly amusing’, but rather that it seems oddly arbitrary and lacking the pointedness or relevance to its context which we might expect in even the feeblest witticism.Footnote ¹⁰ It is certainly possible for humour to be generated from the antithesis of uir and homo. At Terence, Hecyra 523–4, Phidippus calls to his wife Myrrina, and when she responds with an interrogative mihine, mi uir? (‘Is it me you're talking to, my husband?’), he replies in turn uir ego tuos sim? tu uirum me aut hominem deputas adeo esse? (‘Is it your husband I am? Do you consider me to be a husband/man or even a human being?’) This is, if anything, an even clearer proof that uiri are a subset of homines, as the adeo shows, and it is on this normative relationship of the two words (in contrast to the anomalous one at Q. fr. 2.10(9).3) that the joke partly depends: if Myrrina does not consider Phidippus a homo, then a fortiori she cannot consider him a uir. However, the reference to this standard notion that one must be a homo to be a uir would have no particular point were it not wittily combined with the context-specific wordplay on uir as ‘husband’ (as Myrrina uses it) and ‘man’ (‘Man? I'm not even treated like a human being!’)Footnote ¹¹

To turn from the humorous potential of the uir/homo antithesis to Cicero's comedic practice elsewhere in his correspondence, it can be seen that he does make literary jokes which, however amusing or otherwise we might subjectively find them, are unquestionably pointed and tailored to the specifics of their context and subject-matter. One example is his witty and context-specific use of the poeta auctor conceit to depict Tigellius as being actually ‘sold at auction’ (addictum) by Calvus’ mimetic lampoon, in the act of doing which he picks up and even elaborates Calvus’ own conceit ‘of writing a poem in the form of an auction announcement … in which he himself took the part of the auctioneer and offered Tigellius for sale’.Footnote ¹² Equally witty and pointed, and with an added touch of doctrina, is his play on the double status of Quintus’ Erigona as both tragedy and woman, mock-lamenting that she was lost on the road through Gaul despite owning a fine dog, a learned allusion to the faithful Maera who led her mistress to Icarius’ body, as well as a jibe at the ineffectual Oppius.Footnote ¹³ The letters are also full of witty and pointed philosophical jokes and allusions, as Miriam Griffin has shown.Footnote ¹⁴ To cite but one example, Griffin argues that Cicero's ironic concern to come to see Trebatius ‘before [he] flows completely from [his] mind’ (antequam plane ex animo tuo effluo) subtly alludes to the Epicurean doctrine of sense-perception by means of εἴδωλα.Footnote ¹⁵ In our passage, on the other hand, we might wonder why the (dubious) antithesis of uir and homo even arises when discussing Sallustius’ Empedoclea. There is no obvious reason why such a poem, whether as a poem or as an instantiation of Empedoclean philosophy, would suggest a play on the antithesis of ‘man’ and ‘human’, let alone one which is unparalleled in extant Latin, where, as has been shown, one cannot be a uir without also being a homo. If an emendation could provide an antithesis which preserved and perhaps even enhanced the humour, but removed Housman's illogicality, and had a clear connection with the topic under discussion, it would have a good deal to recommend it.

We have already noted how one of the more obvious antitheses of homo is ‘god’. Among the most famous, or notorious, aspects of Empedocles’ doctrine was his claim to be a god and no longer a mortal. The claim is most clearly preserved in the proem to the Katharmoi (DK B112.4–6):

That this doctrine was familiar in first century b.c.e. Rome is clear from Horace's explicit comment (and partial translation) at the climax of the Ars Poetica (‘while Empedocles wanted to be considered an immortal god’, deus immortalis haberi | dum cupit Empedocles …) and Lucretius’ all-but-explicit reference to the poems of Empedocles’ ‘divine breast’ (diuini pectoris) so that he ‘seemed created from scarcely human stock’ (uix humana uideatur stirpe creatus).Footnote ¹⁶ Noting this connection, Murley suggests ‘a jest … at the expense of Empedocles as well as Sallust’ and unpacks the implications of homo as ‘“But if, in the few days before your return, you shall have read Sallust's Empedoclea I'll regard you as a hero, but like Empedocles not a human being”’.Footnote ¹⁷ Murley's interpretation is attractive, but the secondary, implicit antithesis between ‘human’ and ‘god’ sits uneasily with the explicit (and problematic) antithesis between ‘human’ and ‘man’.Footnote ¹⁸ The most economical solution would be to remove the latter antithesis and the make the former explicit.

One solution which would satisfy all the requirements which we have set so far would be to emend the paradosis uirum to a word meaning god, most probably either deum or diuum. The juxtaposition of forms of deus and homo is extremely common in Latin, and occurs eighteen times in Cicero, albeit more frequently in the plural.Footnote ¹⁹ Of course, for a double entendre to work, there must be a primary as well as a secondary meaning. The playful allusion to Empedoclean doctrine would be clear but there must still be an independently comprehensible sense in which Marcus can call Quintus a ‘god’, even if the allusion grants him a degree of licence to stretch common usage a little. Curiously, diuus does not seem to have been used metaphorically of mortals with superhuman qualities, despite (or perhaps because of) its specific connotations of a deified mortal or an intermediate being between god and mortal, and of course its later use as the designation par excellence of apotheosized principes.Footnote ²⁰ There is far more evidence for the use of deus in this sense, ‘de homine … virtute aliqua praedito’, including numerous examples in Cicero's speeches, letters, rhetorical and philosophical works.Footnote ²¹ Of particular relevance to our passage is the assertion by Cicero's Crassus that the godlike orator is one who does not merely use correct Latin but speaks ornate (De or. 3.52–3): si est aliter, irrident, neque eum oratorem tantummodo sed hominem non putant; … quem deum, ut ita dicam, inter homines putant? (‘But if it is otherwise [than that he speaks correct Latin], they laugh at him and think him not only not an orator but not even a human being; … who do they think, so to speak, a god among mortals?’) Even with the qualifying ut ita dicam, it is clear from this passage (and others where there is no such qualification) that Cicero could use deus to designate a human who excels in some field or other, and did so on occasion in antithesis with homo.Footnote ²² As suggested above, the allusion to Empedocles (and perhaps, as I shall argue below, to Sallustius) and the humorous context would help to justify a slight extension of the usual usage whereby the act of reading a poem ironically reflects superhuman qualities, whether of endurance or discernment.Footnote ²³ It might even be possible that a rare use of diuus in this metaphorical sense could be justified by a verbal echo of Sallustius, but Ciceronian and other Republican usage would tend to point towards deus.Footnote ²⁴

As for how such a corruption could have come about, a misreading of diuum as uirum might seem easier than that of deum, but forms of d and u are not normally alike, and the cause here is far more likely to be psychological. The form could have been assimilated to the nearby hominem, or we might see the metamorphosis of god into man as an instance of polar error, where a scribe writes the opposite of the word he is copying.Footnote ²⁵ This type of corruption is not uncommon in Ciceronian manuscripts.Footnote ²⁶ Cicero's plea at Rosc. Am. 12 that the presiding praetor Fannius ‘avenge the misdeeds with all zeal’ (ut quam acerrime maleficia uindecetis) became, in Naples IV B 17, a paradoxical desire that no good deed should go unpunished, as the scribe wrote beneficia for maleficia. Likewise at Mur. 73, according to the copyist of Venice, Marc. lat. xi. 39 (3929), the public attributes Sulpicius’ laying of charges against Murena for having escorts and giving voters meals and spectacles, not to his excessive zeal (in tuam nimiam diligentiam) but to his lack thereof (neglegentiam).Footnote ²⁷ That a copyist could likewise write uirum for deum is entirely feasible. Alternatively, with either deus or diuus, a devout Christian scribe might – consciously or unconsciously – have baulked at Cicero's apotheosis of his brother in such a context and – again consciously or unconsciously – emended the offence away.Footnote ²⁸

There remains the question of whether Cicero is alluding to Empedocles alone or to Sallustius’ poetic depiction of him. As noted above, Murley sees the joke as being ‘at the expense of Empedocles as well as Sallust’. It is certainly possible that the play on god and man is an allusion directly back to the Katharmoi. Sedley has convincingly argued that the proem of Lucretius’ De rerum natura not only imitates Empedocles’ proem but is meant to be recognized as so doing, and thus assumes familiarity with the latter among late Republican litterati.Footnote ²⁹ Even Sedley, however (incidentally using our letter as his principal evidence), allows that such familiarity could come ‘either through direct acquaintance or through Latin translations and imitations’, including Sallustius. None of Cicero's allusions to Empedocles in the philosophical works are noticeably oblique or seem to assume much prior knowledge, though the reference of his Laelius to ‘a certain learned man of Agrigentum’ (Agrigentinum … doctum quendam uirum) could conceivably be taken as allusive as well as faux naïf.Footnote ³⁰ In considering Cicero's allusive practice in the letters, we might compare the witty allusion to Quintus’ Erigona discussed above, which cannot possibly have referred directly to the text of a tragedy which Marcus never had the chance to read, and hence must look to the original myth (and possibly the wrong myth at that), perhaps as narrated in Eratosthenes’ epyllion. However, in the case of our letter, where we are dealing not with a lost text but one with which both correspondents have some familiarity, it is surely more likely that Cicero is alluding not – or not onlyFootnote ³¹ – to Empedocles directly, but to Sallustius’ poetic rendering of his doctrines and perhaps even his poetry.

If Sallustius’ Empedoclea included a Latin version of DK B112.4–6, it is not improbable that it might have occurred as early in the poem as those lines are in the Katharmoi, and hence be recognizable even by those who had not read it in its entirety.Footnote ³² It is also quite likely that θνητός would have been translated as homo (though mortalis is an obvious alternative possibility) and θεός by either deus or diuus. In favour of diuus, we might note its strict distinction from deus as referring to a minor deity (equivalent to the δαίμων which Empedocles elsewhere claimed to be) or even more specifically to a deified mortal.Footnote ³³ On the other hand, the phrase deus immortalis is not only an obvious way to render θεὸς ἄμβροτος, and far easier to fit into hexameters than diuus immortalis, with its initial cretic in the nominative and tendency to elision or hiatus in other cases, but nicely corresponds to the existing common Latin iunctura, di immortales, of which incidentally Cicero is particularly fond.Footnote ³⁴deus immortalis is also the phrase used at Ars P. 464 (quoted above) to render θεὸς ἄμβροτος and it is tempting to speculate that Horace too is alluding not only to Empedocles, but to Sallustius’ Empedoclean poem. This, of course, can only be speculation in the absence of any other trace of the poem, but it is far from improbable. Della Corte has argued for the influence of Sallustius’ Empedoclea on the speech of Pythagoras in Metamorphoses 15.Footnote ³⁵ If Ovid could integrate such allusions into his depiction of a different philosopher, albeit one with some doctrines in common, it is hardly less likely that Horace, about twenty years earlier, could allude to Sallustius when referring to Empedocles himself.Footnote ³⁶ If Horace is indeed alluding to Sallustius, then this might constitute one further argument in favour of Cicero's writing deum when also alluding to the Empedoclea. However, the argument does not stand or fall on the issue of Horatian allusion.

To sum up, I suggest that Cicero wrote to Quintus deum (or possibly diuum) te putabo, si Sallusti Empedoclea legeris; hominem non putabo. In doing so, he would certainly have alluded wittily to Empedocles’ claim to be a god and no longer a mortal at DK B112.4–6, and probably to Sallustius’ own Latin rendering of that claim. Emended thus, the antithesis does not require the special pleading which has been made for uir/homo and it has specific and pointed relevance to the poem under discussion. It is a matter of taste, of course, but it might also be a little more than mildly amusing.Footnote ³⁷