Coleridge on Classical Prosody: An Unidentified Review of 1797

Some time in 1828 Coleridge drafted a letter to the editor of the Quarterly Review, inquiring whether a couple of articles he had in mind ‘would have a sufficient chance of finding admission into the Quarterly, to encourage me in writing it’.  The draft has no address, date, conclusion, or signature.  Whether a fair copy was ever sent we do not know: if it was, it was no more successful than his earlier and even more tentative attempts to follow in Southey’s footsteps.  But in the preamble he gives an account of his two brief excursions – in 1797 and 1808 – into the ‘immoral Act’ of reviewing.  Only the earlier part of the story concerns us here.

Soon after by occasion of a Scheme ˹or Bladder˺ or Fancy bubble, to the bursting of which the World owes the Thalaba, Curse or Kehama, Don Roderic – in short, Robert Southey, I had quitted Cambridge, and from Opinions which less than two years sufficed for me to outgrow, I had given up all my then very flattering Prospects in the Church, and – married! – I was engaged, and if I recollect aright, thro’ the mediation of Sir James, then Mr James McIntosh to write for the Critical Review – and I wrote an article on Lewis’s Monk, and another on Bishop Horsley’s Tract on the Greek Metres, which were perfected into Print.  But I likewise had written some half a score or more of what, I thought, clever & epigrammatic & devilishly severe Reviews, from a single sentence to the quantum of half a page on sundry Fungi of the Press that had been sent to me, ˹for˺ to abide the operation which united Trial, Verdict, and Execution – but a Remark made by Miss Wordsworth, to whom I had in full expectation of gaining a laugh of applause read one of my Judgements occasioned my committing the whole Batch to the Fire/–.[1]

Coleridge did not in all details ‘recollect aright’.  He did not meet Mackintosh until after he had accepted the Wedgwood annuity at the turn of the year 1797-8 – the annuity which was intended, among other things, to preserve him from the ‘warping of the intellectual faculty’ incidental to journalistic writing.  It was to Daniel Stuart, owner and editor of the Morning Post and relative by marriage, that Mackintosh gave Coleridge an introduction; that was in 1798, and Coleridge neglected to call on Stuart on his way to Germany in September.  But the rest of the 1828 statement is accurate enough.  He did review Lewis’s The Monk in the Critical Review.  He might well forget or not take into account three other reviews of Gothic romances in the same periodical: one in 1794 which consists almost entirely of quotation, and two in 1798 both of which are short and unimportant.[2]  The review of The Monk, embodying the first sketch for his doctrine of Dramatic Illusion and giving the first public evidence of an acute critical power, was worth remembering.  And in the same (February 1797) issue of the Critical Review that carried that article there appeared the other article that Coleridge remembered – a long notice of an anonymous essay On the Prosodies of the Greek and Latin Languages (1796).

A month after these reviews had appeared Coleridge told Bowles that he had been ‘an hireling in the Critical Review for these last six or eight months’ and that he had ‘been lately reviewing the Monk, the Italian, Hubert de Sevrac, etc., etc.’[3]  On the occasion of a favourable review of Poems on Various Subjects; in the previous July Lamb had inquired with pleasant malice: ‘Are you not connected with the Critical Review?’ – but Coleridge’s answer is not preserved.  In any case it is clear that the 1794 review of The Mysteries of Udolpho was an isolated contribution before he was regularly engaged.  And when he told Poole in December that ‘I receive about forty guineas yearly from the “Critical Review” and the new “Monthly Magazine”’ he was indulging in a forecast of prospects rather than giving a statement of income received.[4]  The letter to Bowles would place his engagement with the Critical Review in September or October 1796; and the Monthly Magazine published poems of his in both those months.  Whether or not he wrote at the same time small hitherto unrecognized reviews for the Critical is matter for conjecture.  But on 25 October 1796 he borrowed from the Bristol Library Society a copy of John Foster’s Essay on...Accent and Quantity (1763) and returned it on 9 November.[5]  This borrowing marks the date that he set to work on his first big assignment for the Critical Review, the article on Samuel Horsley’s anonymous pamphlet on classical prosody.

Although the borrowing of Foster suggests that he started with vigorous intentions, he did not complete the review at once: indeed, by 8 December he had lost the review copy and sent the following instructions to John Colson in Bristol:

...the moment you receive this letter, if possible,...buy for me a pamphlet entitled ‘an Essay on the Prosodies of the Greek & Latin Language’ – dedicated to Lord Thurlow.  It is a three shilling touch, I believe.  I had to review & have lost it – & it is of the utmost consequence to me to have it directly.[6]

Colson seems to have acted promptly, for Coleridge again borrowed Foster from the Bristol Library on 13 December.  It is only fair to say that during the last weeks in Bristol he was seriously agitated by financial anxiety and the fluctuating plans for his move to Nether Stowey.  He must have taken Foster’s Essay to Stowey with him when he moved at the end of the year, and did not return it until 9 March 1797 after the librarian had requested its return; and the chances are that the review of Horsley’s pamphlet was completed, if not entirely written, in Stowey and not in Bristol.  So, after some three months, this article was printed in February 1797.  One supposes that its companion piece on The Monk was a less leisurely performance.[7]

The Horsley review gives a vivid glimpse of Coleridge at twenty-four, when he had suffered his first bitter political and personal disillusionment, when he was working on Osorio and preparing a second edition of his poems in the hope that although he was ‘not fit for public life, yet the light shall stream to a far distance from my cottage window’.  He comes suddenly to life in the judicious tone of self-assurance, in the declamatory italics, the rhetorical questions, and the two footnotes – one punitive and the other learned though borrowed.  The reference to Lord Thurlow at the end is ambiguous; but he could not resist a mild pun at the expense of the bishop whose controversy with Priestley was matter of personal concern to himself.  In this article he shows that he has mastered the reviewer’s trick of making the most impressive use of materials conveniently at hand.  But the review throws a more favourable light on Coleridge.  He was able to find and make full use of an appropriately weighty authority – in this case Foster – to which he extends a critical interest as lively as he devotes to the text under review.  Also he contributes a couple of arguments of his own (p. 245, II. 35-41, and p. 246, II. 26 ff.), which owe their force to their good sense and his ability to make connexions between the recondite and the real.  And his final quotation from Foster is an early assertion of the critical principle stated in Biographia Literaria and elsewhere: ‘It is a maxim with me, always to suppose myself ignorant of a writer’s understanding, until I understand his ignorance.’  But those who rejoice in the inconsistencies of genius will be pleased to notice that, despite his Christ’s Hospital training and his unquestionable mastery of Greek, he usually avoided (except when transcribing from a text) the intricacies of Alexandrine pedantry by omitting all accents from his Greek, and was wilfully inaccurate in his use of breathings.  The divergences from Horsley in pointing the Greek in this review are characteristically Esteesean and need not be attributed to careless type-setting.

It is surprising perhaps that Coleridge should have ‘given so long an account’ of an unimportant anonymous pamphlet.[8]  The reason might be youthful punctiliousness in a task freshly undertaken or the acknowledged desire to be ‘clever & epigrammatic & devilishly severe’; but more probably it reflects his increasing preoccupation with the minutiæ of prosody.  The evidence for his keen technical interest, as time went out, becomes voluminous.  And his daughter Sara, who could only speak from what Southey and Wordsworth had told her and from the Highgate years, told a friend in 1844 that her father, ‘whose versification has been greatly admired by critics, was fond of talking about anapæsts and iambuses; and if people admired “Christabel”, as it were, by nature, he was never easy till he had put them in the way of admiring it more scientifically’.[9]  Within a year of writing this review, Coleridge had made a number of poems which exhibit a flawless sense of the values and sounds of words dynamically modulated in the context of verse.  This was achieved, as Wordsworth was not reluctant to record, not by instinctive flair so much as by painstaking labour and ruthless self-criticism.

The most noticeable marks of Gothic romance in the first version of The Ancient Mariner are to be seen in the excesses of horror, overstatement, and archaic diction which Coleridge refined away in later revision.  The great debt he owed to M. G. Lewis, however, was not for these external mechanisms, but for the ‘specific overpowering tune’ of a ballad appended so The Monk and for a ‘pretty little ballad-song’ introduced into The Castle Spectre – a poem peculiar for an ‘innocent nakedness’ which Coleridge himself despaired of attaining.[10]  Gothic romance was important to his annus mirabilis because of such marginal acquisitions and the critical thought they aroused in him.  Bishop Horsley’s pamphlet gave Coleridge nothing directly; but by a fruitful coincidence it brought him in touch with Foster’s Essay, and by setting him to think more specifically upon matters of prosody probably helped to crystallize his earlier more diffuse notions of English prosody at the time when he was coming to full poetic stature.

Critical Review, XIX. 139-44 (February 1797)

On the Prosodies of the Greek and Latin Languages.  8vo.  4s.  Sewed.  Robson.  1796.

This is a very learned essay, with the least possible ostentation of learning.  The first four pages, in a clear and satisfactory manner, comprise the substance of Foster’s two first chapters, namely, the difference between accent and quantity, and the argument that accent does not given quantity in other languages, although it almost always does in the English.[11]  The author proceeds to state the signification of the three accentual marks; the general laws of accentuation among the Greeks; and in what respects these differ from the Latin rules; he then points out the superiority of the Greek over the Latin system, and concludes with insisting on the advantages of reading the Greek language by its own accents.  Inclusively he endeavours to prove, first, that the present marks are faithful notations of the ancient tones, – and secondly, that the marks themselves are of very high antiquity.  On all these points his remarks are ingenious, and for the greater part original; and, with the exception of the last hypothesis, solid and convincing.  Respecting the signification of the three accentual marks, he supposes the acute to be in truth the only accent or tone, the grave being merely a negation of acuteness, and the circumflex nothing more than a compound of the marks of the acute accent (ʹ) with the marks of long quantity ( ̑ ).

‘It was probably originally expressed by the two strait lines joined together thus ┘; and this stiff mark was changed into the curve ˜, partly for the convenience of writing expeditiously, and partly, perhaps, for the greater elegance of the shape.  Long syllables only could bear a circumflex; and this compound mark expressed, that the syllable was to be pronounced both with a sharp stroke, and a lengthened sound.  It was of great importance, that this circumstance should be suggested to the reader’s attention by a distinct mark; because the natural tendency of the acute accent, contrary to the prejudice of the English ear, is to shorten the time of the syllable on which it falls;[12] especially of the last syllable of a word, and of the penult, the two seats of the circumflex.  The reader, therefore, was to put upon his guard, when the acute tone fell upon either of these syllables being long, not to suffer any acceleration of his voice, a natural, but by no means a necessary effect of the acute accent, to take place in violation of quantity.’ P. 9.

The superiority of the Greek to the Latin system of accentuation in reading the Greek language is, thus stated –

‘The two opposite rules for the accentuation of words of more than two syllables, the Greek rule requiring an acute upon the penultima, whatever might be its own quantity, when the final syllable was long; and the Latin rule forbidding the penult to the acute, when itself was short; seem to have been both in some degree arbitrary; since neither was positively inconsistent with quantity.  The rule of the Greek language, however, was much the best considered of the two; as it was the best calculated for the preservation of the true rhythm, with ease to the speaker.  This will appear to trying the effect of both systems in Greek verse...

                           ...‘Ούλομένην, ή μυρί’ ’Αχαιοίς άλγε’ έθηκε.’[13]

‘He who, with the Latins, shall say ούλομένην, though he will preserve the brevity of the two syllables -λο- and -μεν-, will find it difficult not to shorten the two long syllables ου- and -μεν; especially the former.  But he who, adhering to the rule of the Greeks, shall say, ούλομένην, will find that without any effort, and almost in spite of himself, he will give the syllables ου- and ην- in their just length.  The same thing might be shewn in innumerable instances.’ P. 19.

We indeed of this country read the Greek and Latin as we read the English, which differs in the powers of the vowels from every other language upon earth.[14]  Our author well describes the metrical havoc which this occasions.  ‘Long is made short, and short is made long; dactyls and anapæsts are confounded; and the former in heroic verse often turned into amphibrachs, cretics, bacchii, and antibacchii.’  To reform this barbarous mode of reading, and to teach the way of giving accent, so as to be note destructive of quantity, but subservient to it, he considers two things only as requisite – ‘first, to give every one of the vowels, and of the diphthongs, its true power, in its proper place; and secondly, to pay a critical attention to the effect of the fundamental rules of accent upon the tones of words in connection, a point which, perhaps, has never been sufficiently considered.’  He describes at length the powers of the several vowels and diphthongs, and points out the usual errors of our pronunciation, and then enumerates ten changes, which he conceives the tone of connected words to have undergone.  These changes formed the laws of modulation.  ‘Accent marked the tone of the solitary word.  Modulation was the effect of accent upon words in connection.’

Dr. Henry Gally, in his dissertation against pronouncing the Greek language according to accents (published anonymously), observes, that ‘nothing would show the absurdity of the modern system of accents more effectually, than to take a piece of poetry, and place the accents according to the quantity which the doubtful and long vowels and diphthongs have in their respective places.  This would cause such a variation in the place of the accents, arising from the different length or shortness of the vowels and diphthongs in their different situations, as would make the modern system of accents quite ridiculous.’[15]

Our author thinks *so differently, that he has actually printed forty lines of Homer, with the accentual marks changed and transposed according to the changes which the quantities of the doubtful and long vowels and diphthongs have suffered from the laws of position; and we certainly are prepared to give our suffrage to the probability of his system.  The accurate recitation of poetry was held in high esteem among the ancients, and seems to have been studied as art.  Now its difficulty could hardly have consisted in the mere chaunting; whereas to watch the changes in the quantities of the final syllables, and transpose the accentual tone according to those changes without error or hesitation, would indeed require much attention and long practice.  If the system should pass unhurt through the ordeal of sound criticism, to read regularly a few lines of some Greek poet according to it, would form, we should conceive, an amusing and useful †exercise for the higher classes in our great schools.  The young men would at least acquire by it the habit of distinct pronunciation, so important in public speaking, but which so few of our public speakers possess.

*Dr. Gally’s treatise is not once mentioned by our author, nor Foster’s celebrated Reply to list. His silence respecting the latter admirable work seems strange. Rev.[16]

†It was a part of a learned education among the ancients—Idem Thrax sex fecit partes grammaticae; exercitatam in accentu lectionem; expositionem, &c. Sext. c. 16.[17]

We shall now proceed to consider our author’s argument in favour of the antiquity of the present accentual marks.  Whether these marks (supposing their antiquity to be incapable of proof) do yet ‘exhibit the true speaking tones of the language, such as were used by the Greeks themselves, when it was a living language, and spoken in its purity’ – this question is stated, but not answered.  Indeed Foster seems to us to have proved the affirmative unanswerably, although that the words ending in αϊ or οι are marked on the antepenult in the present system of accentuation, appears to us a suspicious circumstance, and to savour of the ‘vile Iotacism’, which began to prevail about the times of Adrian and Antoninus.  The present essayist has chosen a ground hitherto unoccupied, and indeed given up by the judicious Foster as untenable.  He endeavours to prove that these marks were in common use in writing from a very early age, before Plato or Aristotle, if indeed the invention of them were not coëval with the first writing of the language.  To prove this, he begins by stating the objection, ‘that the marks of the Greek accents are not to be found in monumental inscriptions, in the legends upon coins, nor in many of the oldest manuscripts.’  Now the assertion from Montfaucon is, that there exist no manuscripts with the accentual marks, older than the seventh century.  If this be true, it seems a difficulty not to be removed; but Wetstein, in the quotation annexed to this essay, asserts that the marks are found in manuscripts older than the sixth century.  Montfaucon, if we recollect aright, mentions the particular manuscript which he deemed the most ancient of the accented manuscripts.[18]  This circumstance should have been noticed, and his mistake, if it be a mistake, detected.  The essayist proceeds to authorities: and the first which he adduces is that of Quintilian.  ‘First, then, it is certain the marks of accent were in use in the time of Quintilian.  For we find all the three, the acute, the grave, and the circumflex, mentioned by Quintilian.’  But how mentioned?  As written? or only as to be sounded?  Of the latter the proofs are abundantly clear; but we cannot find a single sentence which could lead us to a conjecture in favour of the former.  But he goes higher.  ‘The marks of accents were in common use in writing in the time of Strabo.’  We have examined the mutilated passage alluded to; and find, as in the former instance, a convincing proof that words were sounded with accents, – and of nothing else.  The geographer says, that the Iliensians, by a transposition of the accent, rendered έπί γούνασιν, ‘for supplications,’ instead of ‘upon the knees,’ in a particular line of Homer; contending that it ought to be έπί γούνασιν.  An old English dictionary now before us, having classed together (in the preface) a number of disyllable substantives and abjectives, as absent, abstract, contract, &c. &c. adds, ‘by the transposition of the accent, these become verbs.’  Would it be a legitimate inference from this passage that accentual marks were in common use in writing in the time of this lexicographer?  And what greater force does the διαοτρεϕειν τας προσωδιας of Strabo possess? – With the same inconclusiveness of argument, our essayist reasons in his authorities from Plato and Aristotle.  The passages, to which he alludes, prove indisputably the use of accentual tones, and make it probable that our accentual marks faithfully represent them; but they prove nothing more.

‘Aristotle, in his Poetics, speaks of acuteness, gravity, and that which is betwixt the two; and, in his Rhetoric, mentions the three accents, the acute, the grave, and the middle.’  These expressions have references entirely to sound, and in no respect to figure.  What Aristotle calls the middle, Dionysius Thrax, a grammarian in the time of Pompey the Great, calls περισπωμενος; and from this word, as descriptive of the figure of the written circumflex, it has been inferred that the marks must then have existed.  If this argument prove its existence in the time of Dionysius Thrax, it certainly militates against its existence in the time of Aristotle.  But in truth the word is not descriptive of the circumflex mark, as it is exhibited in the other accented manuscript.  Magliabecchi informed Wetstein, that in all the most ancient MSS. the circumflex bore the form of an inverted v. ‘Circumflexus ^v inverse formam ubique refert.’[19]  We would render περισπωμενος, by ‘drawn out in rounded tones,’ in which sense it would indicate the sound only of the accent, in the same manner as its two companions, the grave and the acute.  His next argument is, if possible, still more weak.  It was an Athenian law, ‘Ετάιρα χρυσία εί ϕοροιη δημόσια έστω,[20] i.e. If a courtesan wear golden trinkets, let them be forfeited to the public.  But if the word δημόσια were accented on the penult instead of the artepenult, the sense would be, ‘If the courtesan wear golden trinkets, let her become public property.’[21]  Our essayist adds, ‘This is a very notable instance of the political importance of accents, of written accents, in the Greek language.[22]  For if this law had been put in writing, without any accent upon the word δημοσια, there would have been no means of deciding between two constructions, either of which the words, in this state, would have equally admitted,[23] and it must have rendered an inexplicable doubt, whether the legislator meant that the poor woman should only forfeit her trinkets, or become a public slave.’  Much pathos is here displayed; and we sincerely sympathise with a puzzled judges and the trembling courtesan.  But unfortunately we have a parallel case in our own country, which takes away all appearance of plausibility from this notable instance.  The English statutes are never punctuated; neither are wills: and no man can have attended a court of justice without having witnessed the disputes, and sometimes the important disputes, which this practice occasions.  Without doubt, the legislators foresaw this; but they saw likewise, that more disputes and greater ambiguities would arise from a contrary practice.  Would the doubtful meaning of an unpointed sentence in the law of Moses prove the antiquity of the Masoretic points?[24]

We shall only add, that if our essayist had succeeded in proving the antiquity of the accentual marks, he would have completely overthrown his own ingenious scheme of modulation in poetry.  As the marks must have been added (except on doubtful words) solely as assistants to right pronunciation, it is not credible that they should have been placed in poetry, so as not to give no assistance [sic], but to bewilder and mislead.  This phænomenon can be explained by the lateness of the invention only. – On the whole, therefore, we cannot but be of opinion that the essayist should have acquiesced in the following sensible remark of his ingenious predecessor.  ‘Many diligent persons have with learning and industry laboured to prove, from passages of ancient authors, and other strong testimonies, that these marks of accentuation were not known to the old Greeks.  And they have, I think, proved it satisfactorily: which yet perhaps they might have done as clearly by a shorter way, I mean by this plain argument, that such helps and directions in the pronunciation of a language of any country, are not requisite in writings, drawn up in the vernacular tongue of the nation for the use of its natives, who must be supposed not to want instruction in that respect.’  Foster on Accent and Quantity, p. 178.

The learned and ingenious essay, of which we have given so full an account, is dedicated to lord Thurlow,[25] and has been attributed to a dignitary of the church.[26]  It certainly possesses that manliness of style, which distinguishes the more important writings of the champion of orthodoxy.  If it has been rightly fathered, it is an amusing coincidence, that old bishop Gardiner (the vigorous defender of the then established church) published an essay on a similar subject.[27]