Golden on Innis and Havelock

In ‘The Literate Revolution in Greece and its Canadian Causes: Eric Havelock and Harold Innis1, Mark Golden (University of Winnipeg) presents an extreme2 version of the theory of the Innis-Havelock relation elsewhere urged by Watson, Babe and Carey:   

… there is no denying the similarities [between Innis and Havelock]. What is more, the two men not only overlapped at Toronto for almost 20 years, they even knew each other. Havelock’s reminiscences about his days in Canada (…) came in a pair of memorial lectures  [in October 1978] at Innis College long after the economist’s death [in 1952]. His appreciation of Innis’s achievement remains among the most compelling. Yet he denies any close association (…) and downplays any influence of Innis’s interests on his own work. (…) Havelock, however, is not content merely to assert his independence of Innis. He implies that the current of influence ran the other way: he delivered public lectures at Toronto on orality in Homer in the early 1940s — perhaps Innis heard them. Communication that passed between the two men after he had left Toronto for Harvard leads him to infer that this was so. He sums up the intellectual relationship — “more slight than some may have supposed” (403) — in this way: “In reading Innis, I discover (…) the contiguity between Innis and myself seems to have been, as much as anything, else, a matter of happy coincidence” (424). I [Mark Golden] am reminded of that old classicist’s ploy, anticipatory plagiarism: “I find my conjecture anticipated in the work of so and so.” Curiously coy as it is, Havelock’s account has adherents. For example, Andy Wernick5 regards Havelock’s ideas on Plato as Innis’s starting point [for his late work on communications]. It is true that these ideas [of Havelock] did not see print until 1963 [in Preface to Plato], more than a decade after Innis’s death.  But Innis knew them long before: a letter from Innis to a friend6 in May 1951 mentions a manuscript of Havelock’s “on the question of the shift from the oral to the written in Greek culture.”7 But all this letter really shows, it seems to me, is that even a scholar as gifted and energetic as Havelock didn’t always get his work out as soon as he hoped. Many (…) passages from Innis’s own publications (…) predate this letter. Indeed, his working papers, the so-called “idea file”, include references to the relationship of oral and written language as early as 1944 or 19458; one of the earliest notes, inspired by Ernst Cassirer, asks “how far the clash of written language with oral creates [the abstract] symbolism [needed for algebra]?”9 (…) He was aware of Plato’s relevance to the topic by at least 1946, and speculated, in 1946 or 1947, that civilization is at its peak as the oral tradition shifts to the written. (…) [Such a] mix of literacy and orality seems to have been a crucial element of Innis’s ideas from the start. Havelock, however, came to an appreciation of its importance only late. In his Preface to Plato (1963), Homer represents primary orality and the eventual prevalence of literacy is a triumph of progress. But some 25 years later, in The Muse Learns to Write (1986), “the epics as we know them are the result of some interlock between the oral and the literate”10, “the Muse … learned to write and read while still continuing to sing”11, “the masterpieces we now read as literate texts are an interwoven texture of oral and written”12. This seems to me to be conclusive proof that Innis did not derive his ideas from Havelock… (151-154)

Golden’s “conclusive proof” turns on timing.  If “the mix of literacy and orality seems to have been a crucial element of Innis’s ideas from the start” of his communications work, in 1946 or 1947, and if “Havelock (…) came to an appreciation of its importance only late”, in 1986 or 1987 — a gap of forty years — then it could not be that the former ideas originally derived from the latter.  

However, Golden was apparently not aware that Havelock’s essay in the 1973 Festschrift for his friend and former colleague, I.A. Richards, ‘The Sophistication of Homer’, had already been given as a lecture on January 31, 1946 in Toronto13 and may well have been in circulation there and at Harvard in typescript. Nor that the crucial element of that lecture had been reiterated in a review by Havelock in January 1948 in the University of Toronto Quarterly:

[Owen] plays down the total effect of that enormous weight of tribal baggage, of lore, precept, genealogy, custom, which the [oral] poet has to drag along in his epic. To Owen, Homer the artist is everything; but Homer the encyclopaedist, the didactic recorder of oral tradition, freighted with catalogues and memories, does not exist. This, it seems to me, actually minimizes Homer’s genius, as though he were able to work within the narrower, more controllable limits of a literate method, a Virgil or Dante or Milton armed with pen, picking his themes with nicety, not a bard operating within the great straggling medium of the [oral] saga. If the Iliad is not only astonishing but unique, it is precisely because a controlling perspective, a single point of view, has been imposed upon the most intractable materials. (E.A. Havelock, review of E.T. Owen, The Story of the IliadUTQ, January 1948, 17:2, 211.)

Then, in his 1950 The Crucifixion of Intellectual Man, Havelock put his appreciation of the “mix of literacy and orality” explicitly:

There was a golden age in Athens, when men as they walked the streets lived in two minds at once, guided by the unconscious heroisms of an epic tradition, yet roused to vivid thought by the science of an awakening intellect. (13-14)

So it is not the case “that these ideas [on orality and literacy] did not see print until 1963” or that Havelock “came to an appreciation of [the] importance [of the mix of the two] only late” . More, Havelock’s general thesis was clearly abroad in both Toronto and Harvard at just this crucial time of the middle 1940s through his lectures, earlier publications14 and typescripts. Hence the anecdote told to John Watson by Ernest Sirluck, later Dean of Graduate Studies at UT and the President of the University of Manitoba15:

At this period there was much discussion among classicists concerning the use of epic poetry as a technique for inter-generational communication of the ‘cultural baggage’ of a non-literate people. Sirluck recalls a stimulating conversation with [E.T.] Owen on this subject, with Innis as a quiet, note-taking witness.  Since Innis had contributed little to the conversation, Sirluck was taken aback to see him that same afternoon borrowing from the library all the authorities Owen had cited. When Sirluck expressed his surprise that Innis should be interested in this area, Innis replied emphatically that he thought the subject was of fundamental importance. (Watson, Marginal Man, 297)

And here is Richards in a BBC Third Programme broadcast in October 1947:16

Professor Havelock has suggested that we may see in Plato’s rejections of Homer the revolt of the writing mind’s mode of apprehension against the pre-literate mind’s other, less abstract and intellectual, ways of ordering itself.

While the staple theory of Innis (followed by Havelock at least since 1930)17 illuminated the concrete dynamics of any society (and therefore also of pre-literate Greece), the notion that information storage performs a central (or staple) organizing role in the formation of psychological and social functions certainly came from Havelock.18


  1. In Daimonopylai: essays in classics and the classical tradition presented to Edmund G. Berry, ed Egan and Joyal, 2004, 143-154.
  2. Golden even suggests Havelock may have committed what Golden calls “anticipatory plagiarism”.
  3. Page reference to Harold Innis: A Memoir, 1982
  4. Ibid
  5. Golden references Wernick’s ‘The Post-Innisian Significance of Innis’,  Canadian Journal of Political and Social Theory, 10:1-2, 1986, 128-150, esp 141.
  6. This was Frank Knight, the distinguished economist who was a young instructor at the University of Chicago when Innis and his wife were studying there just after WW1.  Knight and Innis became friends then and remained friends until Innis’ death 35 years later. As he reported to Innis, Knight was asked to play a role in the attempt in the 1940s to lure Innis from Toronto back to Chicago. Knight declined to do so, other than telling Innis that he would certainly welcome him if he decided on the move.
  7. Quoted from a letter of Innis from May 21, 1951 to Knight, as cited in Graeme Patterson, History and Communications (1990), 65.
  8. Golden plainly takes the dates assigned to the ‘idea file’ as reliable.  But their editor, William Christian, specifically warns against this.  He notes: “The advantage of this form (= chronological order) was that it would allow the reader to see development in Innis’s ideas and concerns. However, and this is a very important reservation, the reader is warned that this chronological ordering is tentative at best. There are some clues such as internal dates, publication dates of books Innis used, and (the date of the) use of the material (by Innis). Indeed some of the sections cannot be dated with any certainty at all. In the absence of more information about the original form of the notes, when they were typed and how they were assembled, the present arrangement must stand as one order among many, though I hope it is a broadly reliable one.” (The Idea File of Harold Innis, 1980, xx)
  9. Golden refers here to William Christian, The Idea File of Harold Innis, 1980, 1.8, p 4.
  10. The Muse Learns to Write, 13
  11. Ibid, 23
  12. Ibid, 101, cf. 124, 126
  13. While there were certainly additions to the lecture as published in 1973, like the allusion to the 1969 moon landing, it is remarkable that the examples of ‘the sophistication of Homer’ detailed in it are also cited in Havelock’s 1948 Owen review.
  14. See Sirluck on Innis, Owen and Havelock.
  15. Sirluck was in the military until 1945 and then left UT for the University of Chicago in 1947. E.T. (Eric Trevor) Owen, longtime professor of Greek in Toronto, died in 1948. This anecdote may therefore be dated with confidence to the 20 or so months between the fall of 1945 and the summer of 1947.
  16. For discussion, see Havelock, Innis and Richards in 1947.
  17. For discussion, see Innis and Havelock – 1930 and beyond.
  18. As Havelock was the first to emphasize, his ideas were, of course, importantly influenced by the earlier work of Martin Nilsson, Milman Parry and his teacher at Cambridge, F.M. Cornford. He came to his ideas, he explained, “after encountering the work of Milman Parry, guided also by a reading of Martin Nilsson’s Homer and Mycenae (1933; for me — EAH — still the classic work on the subject), and following (…) intuitions born of (my early) pre-Socratic studies (with Cornford at Cambridge)…” (The Muse Learns to Write: Reflections on Orality and Literacy from Antiquity, 1986, 17)