“Sympathy for fragments?” (USC grad colloquium keynote, April 14th 2023)

Text of a keynote lecture given at the USC grad colloquium, “Fragmentation and Restoration in Antiquity and Beyond” (April 14th-15th 2023) . Thank you to Matteo Barbiero, Joshua Allbright, and Claire Mieher, for organizing the colloquium, and to all of the participants.

I want to begin with a recognition – and fair warning – that some of the themes in the following paper may occasionally be difficult or challenging for some to hear. As will become clear in due course: I have an increased interest in investigating how the concept of the “fragment” relates to broader processes of knowledge, but also to human life and therefore also to violence, decay, loss, and grief. While the majority of this paper takes the form of familiar critiques of philological discourses – and their tendency, in some ways, to trivialize via literalizing processes – there will also be some rawness relating both to bodily perception as well as to several kinds of epistemic and physical injustice. To be explicit and let you know what’s coming: there will be brief discussion of cultural chauvinism (especially Islamophobia), references to incomplete body with resonances of bodily harm in relation to the Venus de Milo, as well as brief mentions of sexual assault in Roman comedy. 

It was about a decade ago that I here – at USC, in Los Angeles – started on my journey with fragments (and at that time, I would have called myself a Ciceronian; maybe, maybe I am now beginning to think of myself as a “fragmentologist” – although I often feel that being a quote unquote “expert” in fragments is like being an expert in nothing; or else that I am willingly entering a kind of void within which I can practice conscious untethering). And, as far as statements of preface go, I do here wish to underscore the fact of feeling as a structural and ultimately generative component to fragmentology – and, indeed, feeling/affect/desire (synonyms but also not-synonyms for “sympathy” in the broadest sense) will, again, broadly (but in multiple modalities) be the content of this keynote today, which I have so frustratingly entitled: “Sympathy for fragments?” Essentially, what happened to me was that I naively entered into a project to study the hundreds of fragments of Latin (and Greek) poetry contained, constructed, and transmitted by the Ciceronian corpus, and emerged from that project with an appetite to understand the fragment as a concept and as (again) a structural feature not only of human text but also of human life. 

For what we call the “fragment” – and what we identify as the processes of “fragmentation” – find their meaning in the context of their use. (Of course.) Within the disciplinary discourses of classics and classical philology, the “fragment” has developed into a rather narrow (yet at the same time, as we shall see, rather mobile) terminological function. Here, we find a certain kind of literalness: the “fragment” (from Latin frangere, “to break”) is the “broken thing.” While it has become a trope of disciplinary discourse to identify this etymology (one which I myself also indulge in), it is so often repeated precisely because it is still meaningful to begin here: to begin with an understanding that what we call “fragment” or “fragmentum” is something that we not only characterize as “broken,” but that we also use faux- or pseudo-classical language – i.e. adopt a Latin term as an assertion of continuity – to describe the phenomenon that we are in the very act of constructing. 

Yes, fragmentum is indeed a Latin word. Yes, it appears in texts from classical antiquity. In the Pro Sestio (79), for instance, Cicero reports that sometime in early 57 BCE Publius Sestius as tribune was attacked by members of Clodius’ gang in the Temple of Castor: “some attacked with swords, others attacked with fragmenta saeptorum,” i.e. “torn” or “broken off pieces of the fence.” Elsewhere, in the De Natura Deorum (2.82), Cicero’s Stoic speaker refers to the earth as a product of intellectual design and not “merely” an inert “clod of dirt” (glaeba) or “lump of stone” (fragmentum lapidis). In each case, Cicero asserts that the “broken thing” is conceptually a “bit” of irrational matter – and other Latin authors write similarly: Livy uses fragmentum to denote broken tree branches (23.24.10), blunted spear tips (32.17.14), and shards of roof tile (34.39.11); essentially products of various moments of collision and destruction.

a piece broken off, a piece, remnant, fragment (class.; mostly in plur.; cf.: frustum, segmentum).

Plur.: inermem atque imparatum tribunum alii gladiis adoriuntur, alii fragmentis septorum et fustibus, Cic. Sest. 37, 79: tegularum, Liv. 34, 39, 11: ramorum, id. 23, 24, 10; for which ramea, Verg. G. 4, 304: avulsarum tabularum remorumque, Curt. 9, 9: crystalli, Plin. 37, 2, 10, § 29: panis, crumbs, Plin. 9, 8, 8, § 25.— Poet.: Emathiae ruinae, i. e. the remains of the army, Luc. 9, 33.—
Sing.: fragmentum lapidis, Cic. N. D. 2, 32, 82.

Despite its apparently ancient pedigree, the term “fragmentum” is one which emerged as a historical concept during later periods of intense intellectual interest in classical antiquity – as a result, the “fragment” is fundamentally a scholarly invention. As Glenn Most (2009: 11) argued, the “metaphor [of the “fragment”] seems not to have been invented until relatively modern times” even though “its foundation was partly laid in antiquity.” That is, even though the word fragmentum appears in antiquity – the idea of the “fragment” (developed out of an interpretation of this ancient term) is fundamentally a modern one. The “invention of the fragment” as a concept is, in my view, well captured by Hans Ulrich Gumbrecht (2003: 2) – writing in conversation with Glenn Most and others – in his definition of one of the “powers of philology”: 

“philological practice has an affinity with those historical periods that see themselves as following a greater cultural moment, a moment whose culture they deem to be more important than the cultural present” 

In Gumbrecht’s conceptualization of philology we see that two points in time are needed as well as an apprehension of relative cultural value: an earlier time of high (indeed “greater”) value, and a later time during which there is an interest in deferential looking backwards. As with all discursive problems, the issue is one of perception as well as rhetoric: philological and fragmentological discourses (to the extent that these have even been distinct modes of inquiry so far) arise during – to quote Gumbrecht again – “those historical periods that see themselves as following a greater cultural moment” which “they deem to be more important.” The idea that the earlier time is a better one is – I put to you – not, in fact, naturally the case, but philological consciousness develops when it is asserted – i.e. thought of and said – that precursor cultures are inherently or innately better than the latecomer-cultures that receive (or perceive) them. 

Putting aside very briefly the question of value, this definition of “philology” as an interrelationship between two points, one prior and one self-consciously “latent,” is still to some degree preferable to the narrowness of limiting the legitimacy of philological inquiry to the “successful” application of a particular set of mechanical tools. For instance, James Zetzel (2018: 3), adducing Sheldon Pollock’s (2015: 22) definition of “philology” as “the discipline of making sense of texts” in fact goes on to assert a history of philology in relation to the ideal of textual correction. Zetzel (2018: 22) writes, for instance: 

“Whatever ‘editing’ [scare quotes] Lampadio (or Stilo or Varro, for that matter) did, it was not textual criticism as we understand it.” 

The early history of Roman philology is here exemplified by Gaius Octavius Lampadio, who reportedly divided Naevius’ Bellum Punicum into seven books (Suet. Gramm. 2); L. Aelius Stilo (the “Pen”), Stoic and teacher of both Cicero and Varro in cultural-historical, philosophical, and textual matters (Cic. Brut. 205-206, Ac. 1.8; cf. Čulík-Baird 2022: 156-157); and indeed, Varro (often termed the “Roman Polymath”) himself. But the history is narrativized as one of deficit shaped by the establishment of a modern ideal; an interesting example of ancients failing to meet the standards supposedly originating with them but rather imposed upon them by modern subjects still seeking some form of continuity. By speaking here (in a rather extended tangent and parenthesis) upon the relationship between philology and fragmentology (again: not really distinct in the discipline – at least not yet), I am repeatedly stumbling upon the fact that we unreflectively use terms derived from antiquity to describe and enact modern discursive positions (from Cicero’s or Livy’s detritus of fragmenta to Plato’s use of “philology” or “philologist” (φιλολογία, φιλόλογος = Tht. 146a, Phdr. 236e) to mean the “love” and the “lover of argument”). In fact, even a brief and non-linear cultural history of such terms serves to verify Gumbrecht’s argument – that philology appears as a form of interrelativity between two historical points. 

Early instances of the term fragmentum (on its way to the meaning of “fragment” as we now use it within the discipline) speak to an essential ambiguity of the term, an ambiguity which captures some of the essential mobility and instability of the fragment itself. For Petrarch (1304-1374), fragmentum did not mean the excerpt of text which had become dislodged from the rest of its body, but instead evoked imagination – in fact, more than imagination, but nostalgic longing and desire – in the face of loss. As Anna Carlotta Dionisotti (1997: 16-17) discussed (and this is her translation), Petrarch describes the ruins of Rome as fragmenta (Fam. 6.2):

et euntibus per moenia fractae urbis et illic sedentibus, ruinarum fragmenta sub oculis erant. quid ergo? multus de historiis sermo erat.

And as we walked around the walls of the broken city and sat there, the fragments of its ruins (ruinarum fragmenta) were before our eyes. So, of course, we talked long of its past.

In the 16th century, Alessandro Guarino reported that he had seen a “large fragment” after line 10 in Catullus poem 2 in an “ancient manuscript” (in codice antiquissimo), leading some to believe that he had seen Catullan verse which had since vanished. But – as Michael Reeve (1996: 22) noted – Guarino’s ingens fragmentum meant “a large gap” not “a large piece of text.” It is precisely this kind of vanishing act which Maurice Blanchot in The Writing of the Disaster (1995: 60) ascribed to the fragment, with its tendency “to dissolve the totality which it presupposes” as well as “to maintain itself as the energy of disappearing.” 

The nature of the fragment as both emergent as well as absent notwithstanding – it was in the 16th century that the term “fragmentum” began to be used in collections of excerpts (i.e. quotations) mined out from citing sources. Anna Carlotta Dionisotti (1997: 24) suggests that one of the first editions of “fragments” preserved exclusively by quotation may have been Carlo Sigonio’s Fragmenta Ciceronis (ed. 1 = 1559), which collected ostensible pieces of the Ciceronian corpus not transmitted by independent manuscript traditions but instead by a variety of excerpting sources. (And this is an interesting case, too, because Sigonio collected “pieces” of Cicero’s De Republica – not knowing that more of it would emerge in Angelo Mai’s discovery of the Vatican palimpsest [Vat. Lat. 5757] in 1819).

Cover plate of Carlo Sigonio's FRAGMENTA CICERONIS
Carlo Sigonio’s Fragmenta Ciceronis (ed. 1 = 1559)
Image: Google books

At any rate, the term fragmentum as a definition for bona fide remnants of ancient texts was regularized in the 16th century and would find repeated use through the 19th century within the philological discourses of German romanticism, at which point it solidified into the scholarly terminology which still sees regular use today. 

(In addition to fragmentum, the Latin term reliquiae was also used – e.g. Johannes Vahlen’s Ennianae Poesis Reliquiae (ed. 1: 1854); or Hermann Peters’ Historicorum Romanorum Reliquiae (vol. 1., ed. 1: 1870); reliquiae being a Latin word which means “remains,” but also, in ancient Rome, referred to more mundane and morbid “leavings,” such as table scraps, the remains of human bones post cremation, as well as excrement and “the (unconsumed) remains of the flesh of a sacrifice” (as reported by Lewis & Short) – in essence: guts, gore, and an unfinished meal scattered around conceptually as if upon the asarotos oikos, the “unswept floor.”

When the term “fragment” is used in contemporary scholarship – that is, today (okay – let’s say within the last decade) – it can come with some self-reflection even while assenting to the premise or rhetoric of fragmentation. For instance, in the general introduction to the massive (the usual word is “monumental”) three-volume, multi-authored 2013 edition – The Fragments of the Roman Historians – issued under the general editorship of Timothy Cornell, the first words of the introduction perform a brief and refreshing moment of disavowal (p3): 

“Only a handful of works on the history of Rome by Romans survive for us to read, as against more than a hundred attested writers whose works are lost. These are known to us only through citations and references in later authors. By a generally accepted scholarly convention the citations are known as “fragments,” some preserved in the form of direct quotations, but many paraphrases of what the original author wrote.”  

Such circumspection quite naturally arises in a volume collecting historical (or rather, historiographical) fragments which honestly might be the most difficult kind of fragmentary genre. It’s either history or oratory. (Poetry at least has metrical structure – and that does keep the bits together in a more agglutinative form.) At stake here is the vexed issue of the difference between fragment and testimonium – we can talk about it, but let’s set it aside for now. The editors here say: we understand that this is a construct, but we’re using it as short hand – as well as a nod to the history of editing that we see ourselves as part of. Likewise, every introduction to oratorical fragments must unpack the difficulty of conceptualizing fragments of speech which originally took the form of an ephemeral utterance regardless of the formality of the immediate oratorical context. Gesine Manuwald (2019: xx) in the general introduction to the Fragmentary Republican Latin Loeb (Vol. 3: Oratory, Part 1) also wrote of fragments in scare quotes: 

“Editing the ‘fragments’ of Roman Republican oratory is in many ways more complex than editing some of the other writers or literary genres represented in Fragmentary Republican Latin (FRL). In addition to the standard difficulties created by any fragmentary corpus (for instance, problems of transmission and attribution), the task of presenting the remains of Roman Republican oratory is confronted from the start with the essential and difficult question of whom and what to include or not to include.”   

There can be an intense – some might even say, absurd – degree of precision involved in the curatorship of fragmentary editions. I defer here to one of my precursors at UCLA, Sander Goldberg in his 2013 review of the 2012 volume of the new Tragicorum Romanorum Fragmenta, edited by Markus Schauer:  

“Readers experienced enough to extract that crucial information from the apparatus are clearly the target audience for TrRF, and so it may be unfair—but remains nonetheless inevitable—to wonder whether so meticulous a presentation of all these scholarly thickets is an act of great generosity or arrant pedantry.” 

(In the end, he actually lands upon the former:

“When faced with a scholarly record on this order of complexity, there is much to be said for the thoroughness we sometimes call pedantry, and while this volume may on occasion make us gasp or sigh, it will continuously and rightfully earn our respect and our thanks.”)

(And remember that I began today saying that there is something fundamentally emotional about fragmentology – an edition might so affecting so as to provoke gasps or sighs! But there more emotions ahead.) In our relationship with what we have invented as the “fragment,” we might (possibly) be approaching a critical mass of deconstructive discourse – one which would fundamentally be preparatory for answering the question of what even we ought to be doing with these fragments now that we’ve found them. (Unless, of course, it is simply enough for the discipline to produce, revise – and continuously iterate – editions of fragments. Which…) 

To the extent that the emergence of the “fragment” as a concept from the 16th to 19th centuries (and thereafter) has formed an object of study in its own right, such interest appears as part of a broad and often euphemistic assertion of the history of quote unquote “European” identity. In her 2022 contribution to the Res Difficiles conference series (which I – alongside Joseph Romero – annually co-organize), Chella Ward noted, for instance, that the much-celebrated 2010 volume – The Classical Tradition – edited by Anthony Grafton, Glenn Most, and Salvatore Settis, laid out the mission of the book as an investment in an exclusive classicism. (This next part will touch briefly upon Islamophobia in connection to other exclusionary discourses.)

They – Grafton, Most, Settis – write (2010: vii): 

“Understandings and misunderstandings of ancient Greek and Roman literature, philosophy, art, architecture, history, politics, religion, science and public and private life have shaped the cultures of medieval and modern Europe and of the nations that derived from them – and they have helped to shape other cultural traditions as well, Jewish, Islamic, and Slavic, to name only these. Every domain of post-classical life and thought has been profoundly influenced by ancient models.”

Upon this statement, Ward comments:

“These short lines attempt to pass as neutral a number of assumptions that are deeply ideological. The list of later cultures ‘classical Graeco-Roman antiquity in all its dimensions’ has shaped masquerades as a neutral description of classical influence – but in fact amounts to a carefully curated exclusion of ‘Jewish, Islamic and Slavic’ cultures from the most classically proximate group, ‘the cultures of medieval and modern Europe and of the nations that derived from them.’ To take only what Grafton, Most and Settis call ‘Islamic…cultures’ as a representative example, it is clear that their exclusion of Muslims from Europeanness in the world narrative they construct here is an Islamophobic fiction. There are multiple Muslim-majority countries in Europe and about 44 million Muslims; Islam has never been ‘other’ to Europe, except in Islamophobic and more broadly orientalising and racist discourses (nor has it, for that matter, been any less proximate to the texts and artefacts of ancient Greece and Rome).” 

Given that what we call the fragment relates not only to textual partiality and – in many cases (for instance) verbal processes of enclosure (i.e. texts quoting other texts, or more broadly, speech acts quoting other speech acts) – this phenomenon is a feature of almost all human expression – globally, planetarily – and is not limited to the formations of canonicity and curation applied with such care to ancient Greece and Rome. To build upon the vector of Chella Ward’s critique: it might regularly be commented, for instance, that Islamic scholarship played a role in the preservation of Aristotelian philosophy (instantiating an example of “fragmentation” – which, after all, regularly entails translation), but there is also a broader and more important point: namely, the fact that Islamic cultures instantiate their own forms of classicism and fragmentation notwithstanding their interrelativity with what has been called the classical world or the classical fragment.

Essentially, in beginning to frame our relationship with what is called the “fragment” within global contexts and intellectual histories – outside of (principally) what I would term an oversympathetic rhetoric of loss, I follow Dan-el Padilla Peralta’s (2022) assertion in “Classicism and Other Phobias” (a lecture series which he presented at Harvard’s Hutchins Center last spring, part of which he also delivered at UCLA Classics last week): 

“the discipline of classics as it comes to life in the racializing, imperializing, and settler colonialist settings of the early modern and modern Euro-Americas becomes overrepresented as the dominant and in time the only mode of classicism.” 

In regards to the fragment itself, seeing that it emerges out of these “European” (and later “Euro-American”) discourses, the history of its development as a concept as – to use Alexis Shotwell’s (2016: 14) language – “coproduc[ing] the age of colonialism” still remains to be written. To speak more plainly: it is not an accident that the concept of the fragment emerged and solidified during a period of accelerating colonial accumulation hand-in-hand with what, in relation to the collections of ancient imperial archives, Dan-el Padilla Peralta (2020: 157) had previously termed the “warehousing of knowledge” as the “deracination” of knowledge. In anticipation of what might be said on this topic in the future (whether by me or someone else) minimally we may say for now that the mining out and isolation of fragments from textual “wholes” in order to exert control and order over an unruly and organic mass of textuality emerged at the same time that such imperial controls reorganized and tabulated the world as a colonial project. Since at least Hayden White’s (1987: 8-9) study of chronicles (“lists”) as a form of narrative which assert their own significance, we should be able to understand that the representation of knowledge even in formal terms – i.e. the fragment as form as it appears on the page – always makes an argument of some kind.       

Indeed, if I were to be more explicit about the nature of my objection (if that is the right word) to the premise of fragmentary editions — then and now — my complaint would address the decontextualizing nature of the fragment within blank, white space. (Although I am far from the first person to notice, comment, or indeed act upon it: Hans Ulrich Gumbrecht (2003: 7) identified a “geometrical dimension [to fragmentary editions], that is, the empty margins around the text”; Anne Carson (1992: 12) noted how scholars used typography to represent uncertainty regarding the text of Simonides as transmitted by Plato’s Protagoras; and Henry Jocelyn (1967, repr. 1969) is a famous gold standard in fragment editing for his representation of plenitude in his edition of Ennian tragedy.) 

For however the fragment is made, it is made by a material holding – and the concept of breakage is to some extent, indeed, to a great extent merely conceptual. Because what we call fragmentary is always in dialogue with something else – not the “original,” no (not in a literalist sense, anyway) but also not with nothing either. In fact, we are the ones who are breaking the fragment. We are the ones who are making the fragment speak to nothing. Even in the case of an inscription or papyrus, there is materiality: substance of stone or paper in addition to earth (and when broken objects are found — they are still found in relation to other objects within archaeological assemblages, as well as in relation to soil and landscape); in the case of fragments made through quotation, there is a textual embrace (cf. Čulík-Baird 2022: 20). The tradition of separating fragment from its frame (much like the contextual void of specimens and artefacts within museum spaces behind panes of glass) is a method of divorce (wheat from chaff) that has a hierarchical function and asserts a possibility of unmediated access. 

There is deep desire here – one that has a mentality of purity. It is a desire for something not as it is – and not as it was, either, but as it could be. We operate here under a metaphysical presumption that piecing together remnants might actually restore something (it conjures some kind of presence, to be sure, but is that restoration?), while that practice of bricolage tends to ignore the contextual plenitudes which bring us the pieces we so admire. Who cares about Cicero, or Aulus Gellius, Plutarch, Athenaeus – or, god forbid, Lactantius??? – when we’ve got mining to do. Yes, Gumbrecht defined philological longing as a relationship between two chronological points – but the classical tradition, mediated as it is through successive quotations of quotations of quotations filtering things in and out and in again (and out again), cannot sustain the conceptualization of a merely binary relationship. Latency is built into paths of transmission – bits and pieces appear downstream. Friends, this is not something to lament. It is simply the way of things. 

But despite what they say – there is nobility in latency. In fact, in latency there is evidence of a continued connection; to some degree, the concept of the fragment is a direct attempt to sever the very connection we have: to sever the relationships between ancient texts out of a perception that one is more important than another. At the same time, one of the lessons that processes of fragmentation can teach us – and this point is one which I take to be a very serious one – is that we, human beings as well as the cultures we make, do not last forever. Our monuments ask for immortality – Sappho fr. 147 says: “someone will remember us in the future” (incidentally, this piece is transmitted by Dio Chrysostom 37.47) – and so far, she’s right, at least in that we remember the words and the sentiment: we aren’t capable of knowing and therefore remembering “us.” While text might be presence, it is not – I think – life. So when we look to what we call fragments, what we are seeing underneath it all is a natural process of decay, not just a process of loss around which to form a rhetoric of desire. We do lament that we have “lost” this text or that text (my rule, by the way, is starting to be: if you can quote from it, it’s not “lost”) – but: I wonder if there’s something important in realizing that we were never going to be able to hold onto everything. And, again, there is nobility in decay. The theory of the “survival of the fittest” is just a theory – and there is no fatedness to the fragment. 

At its core, the fragmentological problem is also a problem of ego, or maybe more the heart – which is where (and we’re this far in now) the “sympathy” comes in. Page duBois (1995: 34) in her critique of Peter Fuller’s 1980 essay on the Venus de Milo (“The Venus and ‘Internal Objects’”) starts to get at this issue. In her engagement with Fuller, she writes that: “Fuller seems to believe…in the transparency of the [Venus de Milo] and to want to make a claim for our unmediated access to it.” She remarks upon his belief that the Venus in its present state relates an “apparent aesthetic superiority” over its “original.” Via its incomplete nature, the statue also “encourages us to complete it internally.” Finally, there is something of the fragmentation which invites the viewer to identify with it. 

Without getting overburdened by (or indeed, assenting to) the weight of psychoanalysis (or its vocabulary – which here does posit the question of bodily harm), there are at least three things that I would remark upon. Firstly, there is the belief that the fragment is aesthetically pleasing or otherwise attractive precisely because it is “broken”/partial/incomplete. This theory for sure bears out elsewhere in the phenomenon known as the “touch of Sappho” – which Nora Goldschmidt (2012: 2) has also identified in the historical attitude to Ennius (the “touch of Ennius”): 

“like the ‘touch of Sappho’ in Victorian England, Ennius exerts a fascination that exceeds that which he would have generated had his text survived complete.”

There is something, then, about this partiality and lack which, in fact, increases desire. We want what we can’t have. Secondly, there is the belief that the “broken”/partial/incomplete object demands completion, and not through physical restoration in this case but via a process of internal completion which entails (in the psychoanalytic terms of Melanie Klein) “internal objects,” i.e. a mental or emotional image of an external object that has been brought inside the self. Through this process, the observer casts their own idealized image onto the fragmentary object – in this case, an idealizing completion of the idealized representation of the feminized human body. An idea of what the human body ought to be like is postulated by the fragmentary object as well as by the observer. Again, without assenting to the premise of psychoanalysis, we can nonetheless note that such a process had been captured by Anne Carson’s theory of readership and desire in Eros the Bittersweet (1985: 145)

“As a lover you want ice to be ice and yet not melt in your hands. As a reader you want knowledge to be knowledge and yet lie fixed on the written page. Such wants cannot help but pain you, at least in part, because…you watch the object of your desire disappear into yourself.”   

What is fragmentary appears to demand some kind of stabilizing or completing intervention – but such an intervention does not necessarily, or rather, necessarily does not arise from the object itself (which is, of course, “lacking” in this way) – and instead the act of completion comes from within the observer. This is simply (or perhaps not so simply) a good explanation for the ultimately rhetorical premise of objectivity – or say, the limits of philological discourse as a scientific endeavour. While it’s often said that textual restorations are in fact quite subjective or even creative, no one can deny – as my grad students said in seminar last week – that such restorations are nonetheless asserted as acts of authority. And yet they are fundamentally sympathetic in the sense of mutually entangled between fragmentary object and fragmenting subject – they come from within the classicist or philologist as well as from the fragment itself. We recall that reliquiae sometimes means “guts” – well, our guts are in this too. Thirdly and lastly, there is the belief that the fragment is an object of desire precisely because human beings – living with a variety of existential and epistemological partialities – see themselves as fragments too. While this is the most abstract and capacious form of sympathy, it is one which ultimately drives many of us to study the ancient world in the first place. Perhaps with the belief that we will ourselves become complete that way. 

And so, there is desire at the heart of the discipline. To some degree, Petrarch’s longing among the “ruins” of Rome was later echoed by Enrica Malcovati (1930: viii) whose Latin preface to the fragments of Roman oratory is here translated by another esteemed UCLA precursor, Amy Richlin (2013: 115)

… cum … ex his quamvis laceris fragmentis, eorum temporum quasi imaginem vitae exprimentibus, et plebis in comitiis fremitum et iudiciorum tumultus atque senatus disceptationum sonitum audire viderer. 

…when, from these fragments that depict, no matter how tattered they may be, a sort of image of the life of their times, I seemed to hear the hubbub of the populace in their assemblies and the shouting of the courtrooms and the boom of the debates in the senate. 

It’s a nice thought – that fragments open up a kind of human hum. What I like most about it, though, is the indication that reading fragments might give you a general or aggregate impression of a culture without also giving you anything particularly specific. And, of course, it is true that sustained engagement with fragmentary materials will expand everyone’s vision of antiquity beyond the overinflated dominance of the “full” canonical texts which take up most of our discursive bandwidth. But there’s more to this, I think. In general, I myself now find such sentiments – permeated with the rhetoric of loss and the sympathy for the fragment to be rather cloying. And beyond my own emotional reaction to the repeated expression of “would that we had more,” I also begin to ask myself what we are missing by epistemologically rooting our intellectual inquiry in such sympathy. For if the purpose of the fragment is simply to be found – and to be appreciated as found – then the fragment is definitionally foreclosed to an analysis of power. In particular, I am compelled by the kind of fragment that contains – and here comes some increased rawness – evidence of domination, power, or exploitation in antiquity but which is in fact sanitized by the processes of quotation which preserve it as well as by the scholarly structures which serve it. 

In seeking out evidence of power in our texts – I of course take inspiration from Amy Richlin’s sustained study of enslaved subjectivities within the Plautine corpus in Slave Theater in the Roman Republic (2017). Among the fragments of Naevius’ Bellum Punicum, for instance, we find the following (and here I am excerpting from the text and translation of the 2022 Loebs by Robert Maltby and Niall Slater): 

plerique omnes subiguntur sub unum iudicium, "almost all are brought under a single judgement"* *=possibly describing the outcome of a council of war or meeting of the senate

Some remarks, then (and some demonstration, at last, of fragmentological praxis – at least in the fine grain). The excerpting author Aelius Donatus (4th c. CE) commenting on Terence’s Andria (55), remarks upon the fact that the first two words are an “archaism” and adduces a line of Naevius as another example. Maltby and Slater in their note suggest that the verse – plerique omnes subiguntur sub unum iudicium, “almost all are brought under a single judgement” – is “possibly describing the outcome of a council of war or meeting of the senate.” Possibly! In cases such as these there really aren’t sure answers (and, of course, it is deeply vulnerable to write translations of fragmentary texts) – but I do have some doubts.

definition of subigo in Latin dictionary + Livy 36.39.3 (where subigo + iudicium appear together)

For one thing, the Latin verb subigo can mean something much more forceful, namely: conquer, subjugate, and subdue – and, indeed, this very verse is adduced in Lewis & Short as part of the verb’s definition of dominance. Yes, iudicium rather broadly means “judgement” – but doesn’t this comparison in Livy (36.39.3) describing the process of Roman conquest, where subigo and iudicium appear together, make a compelling case for something more violent? The Rhetorica ad Herennium (1.14), a Roman handbook designed to teach orators the basic components of their craft, quotes verses from a now unknown Latin comedy (Ribbeck inc. inc. 3-4) as an example of what not to do; orators should not repeat themselves (and: warning here of the language of sexual assault so prevalent within Roman comedy – indeed a set of verses which benefit from Amy Richlin’s vector of analysis). Text and translation is Harry Caplan’s 1954 Loeb

et ne bis aut saepius idem dicamus cavendum est; etiam ne quid novissime quod diximus deinceps dicamus, hoc modo:

Athenis Megaram vesperi advenit Simo; 

Ubi advenit Megaram, insidias fecit virgini; 

Insidias postquam fecit, vim in loco adtulit.

Furthermore, we must guard against saying a thing more than once, and certainly against repeating immediately what we have said already, as in the following:
“Simo came from Athens to Megara in the evening; 
when he came to Megara, he laid a trap for the maiden: 
after laying the trap he ravished her then and there.”

The author of the Rhetorica ad Herennium is only interested in these verses of Roman comedy for their perceived oratorical fault – i.e. for the repetition: Megaram advenit, advenit Megaram; insidias fecit, insidias fecit – but, of course, has no issue with the sexual politics of the verses themselves. Insofar as this set of verses has received scholarly attention, it has either been merely to assent to the Rhetorica’s interpretation (i.e. “these verses are repetitive” – obviously, the repetition is a deliberate poetic effect!!), to speculate regarding their position in the original play, or else simply to attribute them to an “anonymous” poet (for Ribbeck this concept was denoted by incerta incertorum). 

Lastly, we might look to Cicero, who transmits not only fragments of poetry but fragments of his own oratorical corpus. In the Orator (232), Cicero quotes from his own Pro Cornelio – the defense of C. Cornelius, tribune of 67 BCE – in order to execute a rhythmic analysis. That the actual content of (what now constitutes) the oratorical fragment matters less than its technical execution is not only argued by Cicero in his self-citation and analysis, but by, for instance, the Loeb (Hendrickson and Hubbell 1938), who put their translation in a footnote – privileging the rhythmic information with their own annotation. The fragment says (Loeb translation): 

neque vestis aut caelatum aurum et argentum, quo nostros veteres Marcellos Maximosque multi eunuchi e Syria Aegyptoque vicerunt

“neither raiment [clothing] nor gold and silver plate in which our ancient heroes, Marcelli and Maximi were outdone by many eunuchs from Syria and Egypt.”

Since Ciceronian oratory as hegemonic text is replete with rhetoric of civilizationialism – the full effect of this statement might fly under the radar for many, but it’s not neutral. Orientalizing contrast between stalwart but ultimately corruptible ancestors and outside forces from Syria and Egypt capable of their moral destruction asserts a position of Roman power in relation to the rest of the mediterranean via the potency of ethnic or, indeed, racial stereotyping. And this example is particularly sharp: this is the only extant use in the Ciceronian corpus of the Latin word eunuchus – which, at the minimum, asserts and constructs a dominant sexual morality of idealized Roman masculinity against an orientalist trope. 

Via this brief examination of Latin fragments – their transmission but also their scholarly curation – we start to get a sense of how limited our tools are when it comes to understanding ancient texts as archives of power. In reframing the textual evidence of antiquity as an “archive,” I of course here follow Saidiya Hartman, whose influential Scenes of Subjection (originally published in 1997) was reissued last year. She writes (2022: 13): 

“I attend to the cultivated silence, exclusions, and forms of violence and domination that engender the official accounts and listen for other sounds, the ways of knowing disguised as jargon and non-sense. The documents, fragments, and accounts considered here, although retrieved for purposes divergent from those for which they were gathered, nonetheless remain entangled with the violence of racial slavery and its afterlife.”

In identifying that her reading of the archive is motivated by a purpose “divergent from those from which they were gathered,” Hartman shows that we do not have to – indeed, in many cases should not – simply assent to the rhetoric of the texts we study, and, in fact, that to do so is to become complicit with the politics of the text. Within the discipline of classics, Emily Greenwood (2022: 353) – in the “Special Issue: Diversifying Philology, Vol. 1” of the American Journal of Philology, which she also edited – warns classicists and philologists not to be merely stenographers of the texts: 

“As philologists we are not stenographers taking down what is in the texts; instead, we should think of ourselves as doing language.”

With “doing language” a citation of Toni Morrison’s (2019: 106, citing her 1993 Nobel Lecture) own exhortation: “We die. That may be the meaning of life. But we do language. That may be the measure of our lives.” Instead, classicists as “philologists” are trapped in the cycles of influence of influence, rhetoric of rhetoric – and within this closed system of sacralization any critique or attempt to leave complicity with the ancient source is framed as an inflammatory call to “burn it down.” It was under the stenographer’s impulse that Enrica Malcovati gathered and examined the fragments of Roman orators and became moved at the emergent hum of their voices – but the pleasure of that resurfacing presence, as well as sympathy for its persistence, to some extent numbs the senses to the broader histories – often archives of pain – which still remain to be told. By virtue of the metaphysical implications of restoration as internal completion, or of mirroring personal experience in perceptions of fragmentation, our guts are now all tied up in the ancient reliquiae. But I put it to you that this kind of mutual entanglement does place us in a challenging position in relation to our deeper understanding of antiquity – one which goes beyond pleasure or sympathy and towards a holistic apprehension of broader structural processes written in the archives of power. THANK YOU 

“The Fragment and the Future” (Swansea lecture, 23rd Nov 2020): audio + text

Text and audio of an invited lecture given at Swansea Nov 23rd 2020 over zoom. Thank you to Ian Goh for organizing the seminar, and to all of the participants. Click here for the handout.


There are two ways that classical fragments are made: they are either (ironically) preserved through an act, or process, of destruction; or they are preserved by enclosure in a later text. The fragments of Sappho are a good and famous example of this process. In the 7th c. BCE, Sappho of Lesbos composed her lyrics; her poetry became so famous that it circulated as a text in nine books, as the ancients report (Suda Σ 107 = Campbell test. 2). But we can’t read Sappho’s poetry in its fullness, the way we read an ancient author whose text was transmitted with a robust manuscript tradition through the middle ages. Instead, what we have are numerous scraps of material, excavated from the sands of Egypt, onto which ancient readers of Sappho had inscribed her words. Sappho is most famous for existing in shards of papyri, but here’s an example of one of her poems which survives today because it was written onto an ostrakon, a broken piece of ceramic:

PSI XIII 1300 (c. 200 BCE) containing Sappho fr. 2. Image: Sailko via Wikimedia (CC BY 3.0)

This bit of matter, whose text was first published in 1937 by Medea Norsa, is the reason why we can read what we call Sappho “fragment 2”, a poem which summons the goddess, Aphrodite, from the island of Crete into the presence of the poet. Significantly, this object was made in the 3rd or 2nd c. BCE — that is, several hundred years after the poem inscribed on its surface was originally composed. Here is another Sappho fragment, a scrap of papyrus whose text, published in 1951, contains parts of the famous poem known as “fragment 31.” This piece of papyrus was produced in the 2nd c. CE — even later than our Sappho ostrakon; indeed, this scrap of Sappho papyrus appeared almost a thousand years after the poem inscribed upon it was originally composed:

P. Oxy. 2288 (c. 200 CE) containing part of Sappho fr. 31. Image: Oxyrhynchus Online

You would be hard-pressed to find a piece of scholarship on fragmentation that does not make reference to the following etymology: the English word “fragment” comes from the Latin frangere, “to break”; therefore, the fragment is, in essence, “the broken thing.” When we contemplate these artefacts of Sappho, brokenness is certainly palpable — particularly in the case of our papyrus sliver, which preserves only a narrow column of Sappho 31. From this perspective, the fragment is not only broken, but broken off from its fuller text — symbolically representing its own partiality. Yet, as I mentioned a moment ago, fragmentation as a process takes place in two ways: the first is preservation through destruction; the second is enclosure, or textual embrace.

In fact, we can use Sappho to demonstrate this second dynamic as well. Returning to our ostrakon: this object gives us the fullest text of Sappho 2, but we actually already had parts of this poem before it was published in 1937, because two ancient authors had quoted her words and embedded them in their own works. Short literary quotations were made by Greek writers in the Roman era: Hermogenes of Tarsus (De Ideis 2.4) in the 2nd c. CE and Athenaeus (Deipnosophistae, 11.463e) in the 3rd c. CE. We can present this information visually:

(By the way, this little timeline is obviously not “to scale” chronologically(!), but it will do.) While Sappho’s text was available, broadly speaking, in antiquity, the way in which it has reached us is historically complex — the text that we have now, partial as it is, is the product of several different ancient mediations: someone wrote out the poem on a piece of pottery in Hellenistic Egypt, but we didn’t find that until somewhat recently; in the meantime, Greek writers in the Roman period copied small excerpts of Sappho into their own works, and as a result, carried Sappho with them along their own path of textual survival. Even though the Sappho we imagine belongs to archaic Greece — the Sappho we’ve got was produced by generations of her readers, many centuries after her own era.

Indeed, the fragments of Sappho that are the most famous — such as fragment 31 (φαίνεταί μοι…; “he seems to me equal to the gods”)— survive more fully than most other Sappho poems because they were preserved by textual embrace; that is, because they were quoted by later writers (again: Greek writers in the Roman era): Sappho 31 was quoted by the author of On the Sublime (10.1) in the 1st c. CE. It is a papyrus fragment of Sappho 31 which I showed you a moment ago; we can present this fragment’s history in visual form as well:

As it turns out, then, examination of Sappho’s textual history demonstrates the extent to which our knowledge of her poetry is mediated by the afterlife (that is: the reception) of her work: it is the fact that later ancient intellectuals became interested in annotating, commenting, and excerpting her work that gives us access to her poetry, even though it is this very history of intellectual practice that is also responsible for certain limitations upon that access.

Crucially, this process which I have just traced in the transmission history of Sappho occurs basically everywhere in the classical tradition. Texts which do not have their own independent manuscript tradition survive for us to read because they have been consciously folded into surviving works by ancient authors. Let me give you a Latin example as well. The works of the 2nd c. BCE Latin poet, Q. Ennius, were deeply influential on the poetry and indeed, the thought, of Roman intellectuals during the Ciceronian period, but Ennius’ works only now survives in fragments. Since material scraps of Latin books don’t survive in the way that ancient Greek books live on in scraps of papyri excavated from Egypt, we rely on the fact that later Latin writers excerpted verses of Ennius, to carry along in their own textual corpus.

Among the most significant excerptors of Ennius is Cicero, who, in the 1st c. BCE, preserved many fragments of Ennius (as well as poetic fragments of several other Latin — and Greek — authors) in his own textual body via the mechanism of citational embrace. Here is a manuscript of Cicero’s De Senectute/“On Old Age” (this 15th c. Italian manuscript is not one of the principals used to establish the text, but it is one which is digitally available; and it’s very nice to look at):

Arundel 124 f89 (15th century Italy) containing Cicero’s De Senectute: British Library

The first words of Cicero’s De Senectute are actually a poetic quotation from Ennius, from his epic, the Annales (Skutsch 337-339; 335; 336); and you can see that a user of this manuscript marked the quotation, writing “Ennius” in the right hand margin. This kind of textual enclosure — voices from earlier times encased in amber by ancient writers — is a fundamental and ubiquitous dynamic of ancient textuality, and in many cases we find ourselves face-to-face with not only one layer of embedding, but a series; enclosure upon enclosure, as intellectuals of each period recursively return to valued texts of the past.


At this point, I basically want to ask: what can we learn about the ancient world, and the way we study the ancient world, by thinking about the fragment — or, by “thinking fragmentarily”? Here are a number of propositions:

1) The fragment by its very nature demonstrates the partiality of our understanding of the ancient world. When we look at a fragment of Sappho, or Ennius, we know that there is something missing, that there is a lack. That lack might have been produced by the randomness of papyrus disintegration; but it is also produced by the selectivity of our excerpting ancient sources: that is, we depend on what Hermogenes of Tarsus wants to show us of Sappho, what Cicero wants to show us of Ennius.

2) As a subset of the first point: the fragment testifies to the partiality of our vision, but it also, inevitably, gestures towards the fact that there were parts of the ancient world — big parts — that were never picked up by texts (and by “text” here, I really mean the ancient artefact in the broadest sense, from text to object to archaeological site) in the first place: the living, breathing parts, the spirit of culture and life which are only accessible or even scrutable to those who experience and contribute to it — the living ancients. The fullness of texts is, in essence, illusory. In reference to this phenomenon, Page DuBois (1995: 53) stated that “all texts are fragments.” That is, texts which survive fully — even if they give the impression of cultural capaciousness, and the possibility of being pressed through to the world which produced them — even these texts present only a partial vision. (This, incidentally, is why our interpretations of antiquity should probably not rely solely on piecing together textual evidence, but ideally will also use a theoretical framework; or at least, not rely on positivism when our sources are demonstrably partial.) When we consider the fact that the ability to produce text in antiquity is dependent upon the writer’s holding of power and social authority, we can appreciate the extent to which so many ancient experiences were not committed to textual form (viz. the relative scarcity of women’s writing; the suppressed narratives of slaves).

Even though fragments often live in the periphery of our discipline (it is only, really, quite recently in the history of the field that they have been studied and theorized as a historical phenomenon in their own right), it is important to remember that fragments, for the most part, represent something that was very famous in antiquity, but which is now difficult to access. Fragments made via enclosure — i.e. because they were quoted by ancient authors — are (generally) preserved precisely due to their authoritative status in antiquity: think of Hermogenes quoting Sappho, Cicero quoting Ennius. For the most part, then, what is contained within an ancient fragment was very famous at the time that the quotation (or enclosure) was made. Fragments interestingly exist as artefacts of inverse quality, depending on what historical point you are able to occupy: to the ancients, the material contained within what would later become a fragment was the most well known; to moderns, the least. Another nice example of this phenomenon is the commentary tradition to the comedies of Aristophanes, which contain a number of fragments of contemporary poets (e.g. the drinking songs of Praxilla of Sicyon; nos. 38 + 41 Balmer = schol. ad Thesm. 528; Wasps 1236); Aristophanes was parodying the most well known cultural touchstones of his day, demonstrating the vitality of the material towards which his plays gesture. If the fragment represents the transmission, via admittedly a complex mechanism, of mainstream information, we are left thinking — what kind of information wasn’t transmitted to us from antiquity?

3) Our understanding of antiquity is mediated, and it is entangled. When we consider the fact of enclosure as a dynamic of fragmentation — that is: when we consider the fact that our knowledge of some parts of antiquity depends on a textual frame — it is easy to be seduced into discarding that frame. In the history of the study of fragments, there have been many editors who wanted to throw away the text which surrounded the fragment: throw away Hermogenes, throw away Cicero, to get to the “pure” nugget within — the nugget of Sappho, the nugget of Ennius. However, in a significant number of cases, the fragment cannot be easily extricated from the framework which preserved it. Anne Carson (1992) has written about the deep entanglement of the fragment of Simonides (fr. 542) in Plato’s Protagoras (339a–346d); and there are many Latin verse fragments in the corpus of Cicero that are so deeply entangled with his prose that it is essentially impossible to separate them. The context of a fragment is, in fact, deeply valuable: not, necessarily, for the reconstitution, or reconstruction of the original, but as a demonstration of the development of ideas — the history of ideas. Again, the fragment forces us to see that our understanding of antiquity is remediated; and it entices us, as well, to value that remediation, rather than ignoring it entirely. As much as we may desire to access these texts in an unmediated fashion, removing fragments from the frames which hold them will not achieve that — and, in the meantime, risks discarding quite useful information about ancient intellectual history.

4) In theory, the fragment should, by its very nature, stand for alterity, heterodoxy, “otherness.” While the contents of fragments represent, to some extent, an ancient mainstream (see: (2)), they challenge modern reconstructions of antiquity which flatten out complex ancient landscapes with narrative oversimplification. Indeed, in important ways, fragmented material presents counternarrative to prevailing thought, especially the back projection of modern identity (straight, white, male) into antiquity: Sappho’s lyric, of course, resists heteronormative sexuality and presents the erotic as a legitimate metaphysic. The modern poet, Josephine Balmer, has explored the heterodox aspect of fragmentation in her recent poetry collection, The Paths of Survival (2017). In this set of poems, Balmer moves backwards in time, starting with the viewing of a papyrus fragment — this one:

P. Oxy. 2256, fr. 55 (c. 300 CE) containing part of Aeschylus’ Myrmidons. Image: Oxyrhynchus Online

— of Aeschylus’ Myrmidons in the Sackler Library, working through each successive moment of rediscovery and loss, back to Aeschylus himself. Aeschylus’ Myrmidons, in which Achilles and Patroclus were explicitly lovers (see: fragments 135-137; cf. Plato Symp. 180a), famously speaks of gay love with the voice of one of antiquity’s most severe authorities. In The Paths of Survival (2017: 22-23), Balmer dramatizes the rediscovery of a papyrus fragment of the Myrmidons, set against the criminalization of homosexuality under fascism in Italy:

“And then, like the first flicker of smoking fire, slow to take,
I found a tattered word: Antelexa:
My heart turned over. I knew it:
speak out. oppose. dissent.

Later there came confirmation
from the professors in Florence;
I had unearthed a precious sliver
of Aeschylus’ lost Myrmidons

a new sigh from a long silence:
A stifled cry shuddering back:
Enough is enough. No more slander,
no more slurs to crush the tongue.
Time now to protest, to dissent

A point of no return. The moment
all the lies might start to shatter”

In the face of fascist neoclassicism, Balmer figures the rediscovery of a scrap of papyrus as a riot; a shattered object which in turn shatters the contemporary construction of reality: in the face of fascist persecution, a fragment of antiquity speaks to the essential legitimacy of homosexuality. This, I think, is the power of the fragment: it presents a challenge. Narratives of the present can be disrupted by fragments of the past.

At the same time, while the ancient fragment disrupts modernity’s conceptions of itself, it does not do so in a manner of simple correspondence. In an Eidolon article from 2018, Sasha Barish reflects upon the extent to which Latin literature reflects the transgender experience. While Barish meditates upon Ovid (rather than a fragmentary text), his words, in my view, reflect “fragmentary thinking”:

“I realized that I loved reading ancient Roman literature around the time I realized I was transgender, and I don’t think the timing was coincidental. I was trans and in high school, and I appreciated Latin poetry not only for its philological beauty but also because it came to me, unabashedly and without explanation, from a different time and from an alien culture. I watched two sets of social rules unfold in parallel. Gender roles existed in the ancient texts, but they were different from what I saw in the world around me. Men fell in love with men, and to some that was nobler than loving girls. Women weren’t supposed to fight in battle, but sometimes they did. Opposite-sex couples had once been individuals existing outside of the gender binary.”

While Barish here demonstrates the power of ancient literature to activate alternatives to dominant narratives, he goes on to make it clear that Ovid’s depiction of gender dynamics is not a one-for-one correspondence to modern experiences. At the same time, Barish recognizes Ovid’s ability to describe a sense of dysphoria which he can relate to, even though Ovid did not intend his words to connote transgender experiences:

“What’s more, I have been able to find a psychological connection between the poem and my own transgender experience—just not in the places I’d expected. The Metamorphoses actually contains some of my favorite pre-modern descriptions of what I’d call dysphoria — psychological, body dysphoria — only it’s never applied to gender.”

That is, even though the fragment can reveal possibilities of human experience that are hidden, sometimes deliberately hidden, from us, it is not simply that the fragment, as a thing that has survived against the odds, testifies to such experiences directly; instead, the fragment may do the work of alterity simply by being a data point that is difficult to deal with, difficult to fit into the narrative of history as it currently stands. We have an option here of saying either: no, this piece of information does not belong with what we already have; or we can say: yes, this piece of information is valuable because it complexifies what we have — places our other texts and objects in higher relief. Part of what I think of as “fragmentary thinking” includes the possibility of expanding our field of vision to include the data point of the fragment alongside what we think we already know, in order to deliberately broaden and complexify the range and depth of our knowledge. The fragment places everything else in higher relief.

5) Fragmentary thinking helps us reconsider the effects of canonization on the wiring of our discipline’s mind. It is, of course, significant that the ancients created a series of canons — that they hierarchized their ideas and their texts; we do not need to erase the fact of those structures. But we can, I think, resist the focus which canonization of texts and ideas produces, to lift our eyes from this tightly defined field of vision. Fragmentary thinking moves us away from the acceptance of canons as natural or neutral by allowing small details to carry as much weight as large structures. Here, I would bring in the work of the 19th c. sociologist, Georg Simmel, whose theory of fragmentation attributed significance to every distinct aspect of human experience, no matter how small. In the Sociology of the Senses (1907: 1026; trans. Frisby 2013: 55), Simmel writes:

“On every day, at every hour, such threads are spun, are allowed to fall, are taken up again, replaced by others, intertwined with others. Here lie the interactions — only accessible through psychological microscopy — between the atoms of society which bear the whole tenacity and elasticity, the whole colourfulness and unity of this so evident and so puzzling life of society.”

While fragments are often conceptualized partitively — that is: as parts of wholes (w-h-o-l-e-s), wholes that are missing — Simmel considered the fragment in abstract as the gateway to the perception of totality. In Sociological Aesthetics (1896; trans. Frisby 2013: 57), Simmel writes:

“To the adequately trained eye, the total beauty, the total meaning of the world as a whole radiates from every single point.”

In such a conceptualization, each individual fragment of antiquity in its own way opens up a portal to totality: from such a perspective, a fragment of Sappho is as important as a Platonic dialogue; a fragment of Ennius as important as Virgil’s Aeneid, principally because it signifies its own position, and because it signifies the relationality of its position. In the meantime, such a distributive theoretical maneuvre allows us to stand back at a distance to observe the contours and the greater shape that each individual piece together presents; to observe the space between things, as well as the space between ourselves and the ancient: in sum, to appreciate a sense of relationality as well as distance.

6) Our understanding of antiquity is not direct; there is a critical distance between us and the ancient world. And that distance is, in fact, data. By virtue of our desire for antiquity, and by virtue of some of our scholarly practices, we often create a false image of antiquity’s unmediated persistence. Take, for example, this image depicting the conservation of a black figure lekythos, recently tweeted by the Boston MFA:

White ground, black figure lekythos (c. 500 BCE) in the Boston MFA (99.526). Images detailing modern repairs tweeted by the Boston MFA.

Such an image, unintentionally, presents a narrative that broken things can be fixed; damage can be healed; wholeness can be restored: antiquity is still here. It makes me think of the end of a poem by Adrienne Rich (“Meditations for a Savage Child II”; 1973: 58):

“these scars bear witness
but whether to repair
or to destruction
I no longer know”

Just as I want to insist upon the value of seeing fragments nested within the texts which transmit them, I also want to insist upon the value of presenting the image of broken things as broken — the image on the far left in this triptych. While these may seem to be maneuvres which pull in different directions, each gesture illuminates the journey that ancient materials have taken — whether that means that they have been damaged and have fallen apart, or that they have been swallowed by a subsequent moment in history. These kinds of meditations allow the distance between ourselves and our objects of study to remain present. Despite the fact that the romanticism of loss is a rhetorical component of Classicism, the fact of that loss, or distance, must remain an essential element of our theoretical toolkit, so that self-reflexivity, and awareness of relationality, inform scholarly analysis.

We sometimes act as though we somehow have a direct line to antiquity, that we have our noses pressed right up against the glass. That we can see the ancient world clearly, without interpretative or cultural bias. Other times, we make universalizing claims about ancient material: telling our students (and ourselves) that Greek tragedy, for example, contains the essence of humanity. Classics as a field has, of course, contributed significantly to the definition of “humanity” as a concept (which is distinct from “humanity” as an entity); the BA that I received from Oxford is not called “Classics” but literae humaniores, “more human studies” — implying, apparently, that by virtue of that education my knowledge of humanity is complete. In the US, the situation is much the same: take, for example, the definition of “humanity” made by the famed Commission of the Humanities (sponsored in part by the ACLS) in 1964:

“The humanities may be regarded as a body of knowledge and insight, as modes of expression, as a program for education, as an underlying attitude toward life. The body of knowledge is usually taken to include the study of history, literature, the arts, religion, and philosophy. The fine and the performing arts are modes of expressing thoughts and feelings visually, verbally, and aurally. The method of education is one based on the liberal tradition we inherit from classical antiquity. The attitude toward life centers on concern for the human individual: for his emotional development, for his moral, religious, and aesthetic ideas, and for his goals — including in particular his growth as a rational being and a responsible member of his community.”‘

The definition of humanities as ultimately deriving from Classics (or at least, Classicism), reveals itself to be woefully inadequate in this very description, centred as it is, firstly, on the needs of the individual (rather than the community); secondly, on the male individual. Humanity as an entity has obviously always been more multi-faceted than the way that “humanity” as a concept is here defined. What we reckon with here is a two-fold problem defined: firstly, by the insistence that the Greco-Roman tradition should be the measure of humanism, and as result, a measure of humanity; secondly, the retrojection of our own values back into antiquity, presenting the relationship between the classical world and the modern world (in specifically this case, the US) as a direct line. The dynamism of the fragment, testifying, as it does, to the complexities, the partialities, and the gaps in the transmission of ancient knowledge, is itself again a challenge: this time to the very possibility of revivifying of ancient values, of bringing them back to life. At the same time, we can also appreciate the fact that the selectivity required to define the concept of humanity so narrowly represents a further stage of fragmentation; testifying, crucially, to the fact that fragmentation is an ongoing process in which we are all still taking part.


And, of course, it must be directly acknowledged that this definition of humanity, dependent upon classical invocation, has had indisputably violent consequences. For example, as Phiroze Vasunia (2013: 240) has discussed, the British colonial presence in India in the 19th century imagined itself as a “civilizing” force by invoking Classics as the precedent for rule: Robert Needham Cust, a British administrator and judge, wrote in his autobiography (1899: 17) that when he “found [him]self” (as though it were some kind of accident!) “helping to rule Millions in their hundreds of towns and thousands of villages” he could not help but think of Aeneid 6.851-8, which, of course, ends with: parcere subiectis (“spare the conquered”) and debellare superbos (“strike down the haughty”). Edith Hall (2013) described the infamous “Rivers of Blood” (1968) speech by the Conservative MP, Enoch Powell — which quoted Aeneid Book 6 (again; 6.86-7) to incite racial hatred against immigrants — as the moment when “classical literature was put to its most shameful use in the history of British oratory.”

Examples of such invocations of antiquity for modern hatred — where elitism meets racism, sexism, homophobia etc.; that is to say, denials of humanity — are plentiful. The consequences of these historical examples are still with us, and fresh examples are made every day. As modern America roils in the face of violence against Black people, typified, but not limited to, the police murders of George Floyd, Ahmaud Arbery, and Breonna Taylor, we cannot forget the notorious anecdote, discussed by Emily Greenwood (2011: 163), which was attributed to John C. Calhoun, who served as Vice President of the United States (1825-1832); Calhoun reportedly measured the humanity of Black people by their knowledge of Classics, saying:

“if he could find a Negro who knew the Greek syntax, he would then believe that the Negro was a human being and should be treated as a man.”

There are some in the field of Classics who might say: these were utterances made by politicians, not by us; this isn’t us. This is where I want to insist upon the lessons that we learn from the fragment. The fragment says: I am made by reception, I am made by recursive return, I am made by invocation. The fragment says: my life is very long and I mutate as I age; my lifespan represents the intellectual history of ancient ideas; my afterlife is threaded through, entangled with, cannot be separated from, the very history which informs your intellectual and social life. The fragment says: more things are connected than you are currently willing to see.

When Ian Goh asked me to speak as a part of this series — which is called “relevant Classics” — I couldn’t help but think: I have heard anxiety surrounding the extent of Classics’ “relevance” since I myself was a beginning Latin student. The question of “relevance” is deeply interconnected with the perception, and the fact, of Classics’ inherent elitism. For a long time, in both the UK and the US, the study of Latin and Greek was deliberately designed as a programme of elite self-signification; put another way: the elitism in Classics was, quite frankly, the point. Even though the circularity of elitism (study Latin to be elite; elites study Latin) is (maybe) slowly coming apart, the structures erected by the history of this kind of thinking, this curricular design, this hierarchy of access, nonetheless remain. This summer, Dani Bostick published an article in AJP — “Not For All: Nostalgic Distortions as a Weapon of Segregation in Secondary Classics” — which stressed the fact of the Latin classroom as a racialized environment in modern America. Bostick writes (2020: 284):

“While racialized barriers to Classics persist, segregation is euphemistically described as low enrollment; systemic cultural failures are blamed on poor marketing, external trends in education, and regressive pedagogy; and proposed solutions to deeper problems often ignore race, as if all students encountered the same barriers.”

Bostick goes on (correctly, in my view) to pinpoint a nostalgia — or desire — for antiquity as a vital mechanism of the segregated classroom. Bostick (2020: 290) notes the consequence of the nostalgia-driven mentality to classical education as a kind of self-mythologizing which facilitates the acceptance of racially segregated environments: Classics is “for” some people, and not for others. If Classics’ relevance has been defined in relation to its usefulness for such self-mythologizing, then it certainly makes sense that its “relevance” would now feel like it is waning. Amongst all of the claims that we could make about why any student should study the ancient world, we have repeatedly chosen to assert the primacy of our own hierarchical relationships — both to each other, and to the ancient material. In the meantime, even though we market Classics as a way to develop historical consciousness, we habitually turn a blind eye to how we arrived here. The field of Classics is ever mindful of its own relationship to the past (or “a past”, or “some pasts”): the ancient past which we study, and the more recent past which established the frameworks of such study. We keep the past and the present in steady dialogue, but what we also need is a futurist mentality. Futurism requires a sober apprehension of our intellectual history, its consequences. Turning away from material circumstances that we don’t think we’ve caused, while still upholding the systems that created such circumstances, is a mistake.

In this context, “fragmentary thinking”/“thinking fragmentarily” is, I argue, a helpful practice, in that it is inherently self-reflexive and inherently deconstructive. It is a mindset which simultaneously apprehends the complexity of the intellectual tradition while minimizing the kind of romanticism which imagines the ancient to be here with us still in a form that is somehow: perfectly present, perfectly whole, and perfectly understood. The fragment testifies the partiality of our understanding of the ancient world; it points to losses on the micro level as well as on the macro level, allowing us to question to what extent our textual picture of antiquity could ever present a totality such as the one that modern self-mythologizing would need as a foundation. The fragment demonstrates the mediated and entangled nature of our ancient understanding: ancient knowledge is transmitted through a number of filters, which testifies both to the ancient tradition’s tendency to loop back upon itself, as well as to the indirectness of our cultural understanding. The fragment stands for alterity and otherness: not only because fragmentary material often contains challenges to what is found in the mainstream, but in the fact of its potential to represent data which cannot be easily reconciled with what we already know. While the fragment itself is, to a significant extent, representative of canonization, fragmentary thinking can help to loose the hold of canonical thinking, within which texts gain greatest weight by virtue of the doctrine of “survival of the fittest.” Finally, and crucially, I think: the fragment, in all these ways, illuminates distance; it communicates the journey which ancient knowledge has taken to reach us, and, as a result, it plants us historically in our own position, here in modernity. THANK YOU.


“Fragment as Form” (UT Austin lecture, 25th Sept 2020): audio + slides

Last week I gave a lecture at UT Austin (via zoom) entitled “The Fragment as Form.” This drew on my recent research on fragments of Latin poetry in the Ciceronian corpus, but also reflected on the ancient fragment more broadly. You can listen to audio of the lecture (recorded earlier that day) here: