Review of Writing Homer. A study based on results from modern fieldwork, by Minna Skafte Jensen 

Gregory Nagy, Harvard University
[This review of Writing Homer. A study based on results from modern fieldwork, by Minna Skafte Jensen (Copenhagen: Det Kongelige Danske Videnskabernes Selskab; The Royal Danish Academy of Sciences and Letters, 2011. 440 S. 16 Abb. Scientia Danica. Series H, Humanistica, 8 vol. 4.) appeared in Gnomon 86 (2014) 97-101. The original pagination of the review will be indicated in this electronic version by way of brackets (“{” and “}”). For example, “{97|98}” indicates where p. 97 of the printed text ends and p. 98 begins.]
The argument of this book is a recapitulation of an argument made already in 1980, according to which the Homeric Iliad and Odyssey first happened in 522 BCE: the historical setting was Athens, in the era of the Peisistratidai. [1] For Jensen (hereafter, I refer to her as “J”, indicating the relevant page number immediately thereafter), these two epics had never been written down before. But then, in the late summer of 522 BCE, they were dictated to an expert team of scribes (J 244). The most prominent alternative dictation theory is the formulation of Richard Janko, who posits the eighth century BCE as the era when the Iliad and the Odyssey were first written down. [2] For Janko, Homer was an aoidos or ‘singer’, and he was a primitive genius. For Jensen, who wants to have the dictation happen more than two centuries later, the genius behind the poetry must be not an aoidos but a rhapsōidos or ‘rhapsode’ (J 229; detailed exposition in chapters 9-10). Actually, there were two such geniuses, one for the Iliad and one for the Odyssey: “I modify the single genius in the sense that I posit one rhapsode as the recomposer of each poem” (J 229). After the dictation, the reception of the text was confined to a small coterie of elites, and this text was “unknown to the broader public” (J 247).
In her efforts to make her argument sustainable, J deems it necessary to reject a formulation that I have developed for explaining the development of the Homeric tradition. I call this formulation an “evolutionary model,” which I built gradually in the course of seven consecutive books that I will hereafter abbreviate as PP (1996a), HQ (1996b), PR (2002), HR (2003), HTL (2004), HC (2009), HPC (2010). [3] My explanation can also be described as a “diachronic model,” since I use the term “evolutionary” as a synonym of “diachronic,” as did Ferdinand de Saussure. So for me “evolutionary” is not the opposite of “devolutionary” (as it is for J 230-237; for more on the term “diachronic,” see HR 1).
In what follows, I outline the diachronic model, utilizing it as a framework for debate with J’s argumentation. From the start, I must stress that this model was built to explain the evolution of the Homeric tradition, not the history of the Homeric texts. Here, then, is the model, divided into five general periods (as summarized most recently in HC 3-5): {97|98}
Period 1 of the Homeric tradition was a relatively most fluid period, with no written texts, extending from the early second millennium BCE to the middle of the eighth century in the first millennium BCE. When I say that this was a most fluid period, I mean that the poetic tradition that eventually became the Homeric Iliad and Odyssey was most susceptible to change in this earliest period of its evolution.
Period 2 was a more formative or Panhellenic period (on the term Panhellenic, see HQ 39-42; not “Panhellenistic,” pace J 214), still without written texts, extending from the middle of the eighth century BCE to the middle of the sixth. Within this period I now reconstruct two sub-periods, 2a and 2b:
Sub-period 2a extends from the eighth to the early seventh century BCE. During this sub-period, the historical context centers on a political federation known as the Ionian Dodecapolis, consisting of twelve cities situated on the mainland of Asia Minor and on two outlying islands located on the Asiatic side of the Aegean sea, Samos and Chios (HPC 22). In positing this sub-period, I am following a model built by Douglas Frame in his book Hippota Nestor (hereafter abbreviated as HN). [4] According to this model, the epic tradition that culminated in the Homeric Iliad and Odyssey took shape within a length of time extending from the late eighth century BCE into the early seventh century BCE, in the historical context of a sacred space in Asia Minor known as the Panionion. This space was the venue for a seasonally recurring festival known as the Panionia, uniting the twelve cities of the Ionian Dodecapolis. From here on, I refer to this tradition of the Homeric Iliad and Odyssey, as it was taking shape at the festival of the Panionia in the eighth and the seventh century BCE, as the Panionic Homer. This Panionic Homer was divided into twelve units of composition / performance, corresponding to the twelve cities of the Dodecapolis, and each one of these units was further subdivided into four sub-units of composition / performance known as rhapsōidiai ‘rhapsodies’, performed by rhapsōidoi ‘rhapsodes’ (HPC 22, following HN ch. 11). These rhapsodies correspond to what we know as the “books” or “scrolls” of the Iliad and the Odyssey. (J 232 “declines” Frame’s model together with mine.)
Sub-period 2b extends from the early seventh century to the middle of the sixth. During this sub-period, the historical context includes not only the festival of the Panionia, uniting the twelve Ionian cities of the Ionian Dodecapolis, but also other festivals, especially the Panathenaia in Athens, located on the European side of the Aegean Sea. The Panathenaia was a festival designed to unite all Ionian cities with Athens as their notional metropolis or mother city (HPC 10-12).
Period 3 was a definitive period, centralized in Athens, with potential texts in the sense of transcripts, extending from the middle of the sixth century BCE to the later part of the fourth.
Somewhere near the beginning of this period, Homeric performance traditions in Athens became definitive in the historical context of reforms undertaken by the Peisistratidai, an aristocratic family that dominated Athens at the time. These reforms, centering on the reorganization of the Athenian state festival of the Panathenaia, culminated in the official adoption of the Panionic tradition of the Homeric Iliad and Odyssey, which as we have seen had been shaped in period 2a, back in the eighth and seventh century BCE. By the time of period 2b, in the sixth century BCE, this Panionic transmission of the Homeric Iliad and Odyssey was mediated primarily by the Homēridai ‘sons of Homer’, a lineage of rhapsodes originating from Chios (HPC 20-28, 57, 59-78). As we have already seen, the island state of Chios had been a member of the Ionian Dodecapolis, and so the Chiote {98|99} transmission of the Homeric Iliad and Odyssey can be traced back as far as period 2a, which marks the Panionic phase of these epics as they took shape in the historical context of the Panionia, the festival of the Ionian Dodecapolis (HPC 28). But then, toward the end of the sixth century BCE, this mediation of the Homeric Iliad and Odyssey by the Homēridai of Chios shifted from Chios to Athens. According to an ancient source (“Plato” Hipparkhos 228b-c), Hipparkhos the son of Peisistratos introduced in Athens a law that required rhapsodes competing at the quadrennially recurring Panathenaia to perform the Homeric Iliad and Odyssey in relay; by having to take turns in performing these epics in the sequence required by the narration, the performers had to collaborate as well as compete with each other in the performances of these two epics (HPC 21-25). The most likely occasion for the first such performance in Athens was the celebration of the festival of the Panathenaia in 522 BCE (HPC 20-21). This is not to say that the Panionic tradition of the Homeric Iliad and Odyssey would have been unknown in Athens before 522 BCE: it is only to say that the tradition of performing these epics in relay was not officially institutionalized in Athens before this time (HPC 68-69). In terms of this reconstruction, the new tradition of performing the Homeric Iliad and Odyssey in relay at the Panathenaia stems from an older tradition of performing these epics in relay at the Panionia, and, as we have seen, the mediators between these two traditions were the Homēridai ‘sons of Homer’, originating from Chios (HPC 69).
In terms of this reconstruction, the Panionic tradition of the Homeric Iliad and Odyssey stemming from the Ionian Dodecapolis in period 2a lasted through period 2b in Chios. Then, in period 3, the Chiote form of the Panionic tradition was adopted in Athens.
This period is linked with mythological narratives that retell how Peisistratos of Athens collected the poetry of Homer, which was left scattered throughout Asia Minor after the poet’s death, and reintegrated this poetry as a totality in his own city (Greek Anthology 11.442; further documentation in HPC 314-318). Such an act of integration, commonly known today as the Peisistratean Recension, is interpreted by some as a historical fact and by others as an antiquarian invention (survey in HQ 93-105). In terms of a diachronic model, however, the interpretation is different: the basic narrative about the Peisistratean Recension is a charter myth, a totalizing aetiology meant to explain the unity of Homeric poetry as performed in the city of Athens: in terms of the myth, grounded in this city, Peisistratos unified Homeric poetry by reintegrating what had become disintegrated in a multiplicity of performances throughout the other cities of the Greek-speaking world, particularly in Asia Minor (HPC 315). There is no need to insist, however, that the entire Iliad and the entire Odyssey were performed on each and every occasion of the Panathenaia. Comparative ethnographic evidence shows that a given group attending a given festival may conceptualize the performance of an epic as a totality that fits the occasion, even if only a part of the given epic is actually performed (HR 15n74).
During period 3, the transmitters of a set of epics known as the epic Cycle were already starting to gravitate away from the tendency of attributing these epics to Homer (HPC 25, 69-70). According to some ancient sources, for example, the epics known as the Aithiopis and the Iliou Persis (or Destruction of Troy) could be attributed to a poet known as Arctinus of Miletus (Proclus summary p. 105.21-22 and p. 107.16-17 ed. Allen); correspondingly, the Little Iliad could be attributed to another poet, Lesches of Lesbos (p. 106.19-20). Such epics of the epic Cycle continued to be performed in Athens during this period, as we can see from signs of Athenian accretions in the content (HPC 320). At the festival of the Panionia at the Panionion of the Ionian Dodecapolis in Asia Minor, on the other hand, such a differentiation between Homer and the poets of the epic Cycle may have been taking place far earlier - as early as the late eighth and early seventh century BCE (HPC 70, 96, 319-320).
Period 4 was a standardizing period, with texts in the sense of transcripts or even scripts, extending from the later part of the fourth century BCE to the middle of the second. It is a misstatement to paraphrase this formulation by saying {99|100} (as J 216 does) that “written texts must be accessible to the bards and their helpers.”
Period 5 was a relatively most rigid period, with texts as scripture, extending from the middle of the second century BCE onward. This period starts with the completion of the editorial work of Aristarchus of Samothrace on the Homeric texts, not long after 150 BCE or so, in the Library of Alexandria.
In the sequence of five periods of Homer that I have just outlined, my reconstruction allows for the idea of Homeric texts - starting already at period 3. (Relevant here is the prehistory of the word text as a metaphor: see PR 98). But I must emphasize that the term text is not specific enough, and that is why my reconstruction features three other terms that are more specialized and thus more suitable for describing periods 3, 4, 5 respectively: transcript, script, scripture. I summarize here my working definitions of these three terms (HC 5):
By transcript I mean the broadest possible category of written text: a transcript can be a record of performance, even an aid for performance, but not the equivalent of performance (HQ 34-36, 65-69). We must distinguish a transcript from an inscription, which can traditionally refer to itself in the archaic period as just that, an equivalent of performance (HQ 34-36). As for script, I mean a narrower category, where the written text is a prerequisite for performance (PP 153-186, HQ 32-34). By scripture I mean the narrowest category of them all, where the written text need not even presuppose performance.
In the course of presenting her critique, J occludes all three of these terms as well as the differences conveyed by these terms. This occlusion is particularly damaging to her argumentation with reference to period 3 of the model, which allows for the possibility that there were transcripts. My use of the term makes it clear that a transcript has no influence on performance. J (217) is simply wrong to claim that “Nagy’s hypothesis attributes to the written transmission features that are characteristic of oral composition and transmission.” In fact, my point is just the opposite: period 3 is a time of oral transmission, not written transmission, and that is why I use the word transcript with reference to any possibility of existing texts. And I do not “imagine” what she describes as “a fertile interaction of performance and written text.” In the same context, she refers to “the dogma concerning the interaction between the two media” (J 217). But I posit no such “interaction,” and that is the point of my using the term “transcript.” Further, the existence of variants in transcripts is a symptom not a cause of multiformity in any oral tradition that is being transcribed. It is a gross misunderstanding to claim that the editors of the Homer Multitext project (Casey Dué and Mary Ebbott) assume that the Homeric tradition was shaped by a process of “a gradual writing” (J 229).
Here is a sampling of other statements that I think are subject to doubt:
J 217: “Plato does not suggest that a victory [for a rhapsode performing] at the Panathenaia would be more important than one in Epidaurus.” (But see HPC 364.)
J 219: “There is no example of alternative narratives, or even of the same story narrated in different ways.” (But see HTL 38-39; HC 590-591.)
J 228: “Even if we accept that the Homeric variants originate with the scribe...” (This premise, attributed to me, is not mine: for me, Homeric variants “originate” from the oral tradition that is transcribed by the scribe.) {100|101}
The last of these three statements that I have sampled goes to the heart of J’s argumentation, as we see from this summation of her views:
I insist on one original for each text [= the Iliad and the Odyssey] and scribal milieus as the home of the variants found in manuscripts.
In response to such a summation, I can do no better than to invoke the words of my late teacher Albert Lord:
Our real difficulty arises from the fact that, unlike the oral poet, we are not accustomed to thinking in terms of fluidity. We find it difficult to grasp something that is multiform. It seems to us necessary to construct an ideal text or to seek an original, and we remain dissatisfied with an ever-changing phenomenon. I believe that once we know the facts of oral composition we must cease trying to find an original of any traditional song. From one point of view each performance is an original. [5]


[ back ] 1. M. S. Jensen, The Homeric Question and the Oral-Formulaic Theory (Museum Tusculanum, Opuscula Graeco-Latina 20; Copenhagen 1980).
[ back ] 2. R. Janko, Homer, Hesiod and the Hymns: Diachronic Development in Epic Diction (Cambridge 1982).
[ back ] 3. G. Nagy, Poetry as Performance: Homer and Beyond (Cambridge 1996); Homeric Questions (Austin TX 1996); Plato’s Rhapsody and Homer’s Music: The Poetics of the Panathenaic Festival in Classical Athens (Cambridge MA and Athens); Homeric Responses (Austin TX 2003); Homer’s Text and Language (Urbana and Chicago 2004); Homer the Classic (Hellenic Studies 36; Cambridge MA and Washington DC 2009); Homer the Preclassic (Berkeley and Los Angeles CA (2010). J refers to all of these books except for the last one.
[ back ] 4. D. Frame, Hippota Nestor (Hellenic Studies 37; Cambridge MA and Washington DC 2009). See especially his ch. 11.
[ back ] 5. A. B. Lord, A. B. [1960] 2000. The Singer of Tales (Harvard Studies in Comparative Literature 24; Cambridge MA 1960; 2nd ed. 2000, with new Introduction [vii-xxix], by S. Mitchell and G. Nagy) 100. The italics are mine.