Back to home page
Back to "Occasional Papers"

Reuven Tsur

Some Remarks on the Nature of Trochees and Iambs 

and their Relationship to Other Metres
(Introductory Page)

download paper
        (pdf file)

This page contains a link (and some updating) to my 1971 paper Some Remarks on the Nature of Trochees and Iambs and their Relationship to Other Metres as well as a bit of personal history regarding the beginnings of my Perception-Oriented Theory of Metre. 

I wrote my first two papers in English metrics in 1971, after finishing my DPhil dissertation at the University of Sussex. The first one was called “Articulateness and Requiredness in Iambic Verse”. It was straightaway accepted (with minor omissions) by Style (published in 1972), in spite of the fact that I was a perfectly unknown entity. I am still grateful to James R. Bennett, who judged my work on its own merit, and not on the merits of my reputation. The second one, called “Toward a Perception-Oriented Theory of Metre”, was repeatedly rejected by journal after journal. Editors said the paper handled too many issues and it should be cut down to a feasible length. This was usually followed by a long list of questions to be answered. By the time I answered all the questions, I was knocking on the doors of publishers with a book-length study, which too was rejected again and again, until Benjamin Hrushovsky pre-published it in 1977, in his Papers in Poetics and Semiotics series. When published at long last, my work was very well received. Terry V. F. Brogan in his English Versification, 1570 - 1980: A Research Guide. Baltimore: Johns Hopkins University Press (1981) characterizes my 1972 article in Style as “An important new approach to the description of verse-structure”; and my Perception-Oriented Theory of Metre as an “attempt to embed the basic Halle-Keyser framework within a much wider, more highly formalized psychological (Gestalt) theory of meter which integrates (1) some of the current understanding of human processing of perceptual patterns and (2) the concept of Performance at a fundamental level in the theory, in order to achieve greater refinements of explanatory power, i.e. greater ‘delicacy’ of description, by converting Halle and Keyser’s dichotomous metrical determinations (metrical, unmetrical) to a continuum of finer discriminations of metrical complexity”. And, also, “the theory goes far beyond the Halle-Keyser framework (Tsur manages to put considerable distance between himself and them in chapter 1). Section 8 on Performance is especially stimulating”.

No wonder, perhaps, that at a time when Halle and Keyser had to struggle for acceptance (see below), a theory that “goes far beyond the Halle-Keyser framework” was repeatedly rejected.

When the paper was rejected by Language & Style, the editor Eddie Epstein invited me to meet him in London, where he was spending his sabbatical. We discussed my theory for several hours. He was especially interested in my discussion of the differences between the iambic and trochaic metre, and the special status of the iambic. He asked me to send him the paper I was writing; but, he added, it will have to undergo the usual refereeing process. I was very much disappointed when he informed me that he had been advised to reject my paper. 

My argument was mainly based on two sets of experiments. One set was H. Woodrow’s tick-tack experiments which found that in a series of tick-tacks, “with equal temporal spacing, a regularly recurring, relatively greater intensity exerts a group-beginning effect, and a regularly recurring, relatively greater duration a group-ending effect” (Woodrow, 1951: 1233). “Intensity has a group-beginning effect: duration, a group-ending effect: pitch, neither a group-ending nor a group-beginning effect” (Woodrow, 1911: 77). In L. B. Meyer’s (1956: 106-107) paraphrase, “durational differences tend to result in ‘end-accented rhythms’, and intensity differences tend to result in ‘beginning-accented rhythms’. [...] Pitch has neither group-beginning nor group-ending effect”. Rice’s more recent experiments (1992) indicate that variations in pitch do lead to a significant shift towards iambic groupings (see below). The other set was D. B. Fry’s experiments in stress perception according to which linguistic stress is cued by a mixture of acoustic cues: pitch, duration and amplitude, in this decreasing order of effectiveness. If pitch differences are irrelevant to grouping direction and duration differences are more effective in stress perception than amplitude differences, end-accented metres should be more natural in poetry in a variety of languages. If variations in pitch also lead to a significant shift towards iambic groupings, it should reinforce this effect. Significanly, even in Hungarian poetry, where stress is invariably on the first syllable, the iambic metre is far more natural and widespread than the trochaic.

I was quite unpleasantly surprised when, a few years later, Language & Style published a paper addressing precisely those questions, relying on precisely those experiments. I said nothing; only in the Preface to my A Perception-Oriented Theory of Metre (1977) I mentioned Robert P. Newton’s essay “Trochaic and Iambic” (Language & Style, Vol. 8, No.2, Spring 1975, pp. 127-156), “which makes a few points uncannily similar to mine in the chapter ‘Some Remarks on the Nature of Trochees and Iambs’”.

Recently I realized that metrists are still puzzled as to the difference between the iambic and the trochaic, and the special status of the iambic, already pointed out by Aristotle and Horace. Since my book A Perception-Oriented Theory of Metre is inaccessible today, I decided to display this paper on my website, as it appeared in that book (with minor corrections).

Since the nineteen-eighties and nineteen-nineties, Woodrow’s work has had an enormous impact on linguistic research, in a very different perspective. To mention only a few studies: The generative linguist Bruce Hayes applied Woodrow’s extra-linguistic principle of rhythmic grouping to iambic and trochaic feet in a wider linguistic perspective:

a. Elements contrasting in intensity naturally form groupings with initial prominence.
b. Elements contrasting in duration naturally form groupings with final prominence.

He was more interested in generalizing to a cross-linguistic theory of stress assignment than in versification, and his work is motivated by in-depth analyses of stress patterns of a large number of languages. (Hayes, B. (1985). Iambic and trochaic rhythm in stress rules. Niepokuj, N., M. VanClay, V. Nikiforidou, & D. Jeder (eds.), Proceedings of BLS 11: parasession on poetics, metrics, and prosody. BLS, Berkeley, pp.429-446. Bruce Hayes (1995). Metrical Stress Theory: Principles and Case Studies. Chicago: The University of Chicago Press. For a critique of Hayes, see Anthi Revithiadou: “The Iambic/Trochaic Law revisited—Lengthening and shortening in trochaic systems”.  In Boban Arsenijevic, Noureddine Elouazizi, Martin Salzmann & Mark de Vos (eds.), Leiden Papers in Linguistics 1.1 (2004), 37-62. Available online:

In Chapter 5 of his doctoral dissertation, Curt Rice (1992) replicated Woodrow’s experiments. “The technological resources for conducting this research are dramatically more sophisticated than those which Woodrow had available”. At variance with Woodrow, Rice (1992:198) showed that variations in pitch do lead to a significant shift towards iambic groupings, a result that, Revithiadou (2004: 38) says, does not lend support to Hayes’s Iambic/Trochaic Law. But it lends additional support to the view expressed here. The reason for this difference appears to be that we have different purposes: Hayes applies the extralinguistic principles of rhythmic grouping to phonological stress rules, while I apply them to versification patterns determined independently from linguistic patterns, as suggested by Halle and Keyser.
Rice, C. (1992). Binarity and ternarity in metrical theory: parametric extensions. Doctoral dissertation, University of Texas, Austin, Texas. Available online: http://www.hum.uit.no/a/rice/v2/writing/rice1992.html

 It all began when I didn’t write a chapter on metrics in my DPhil dissertation. In the years 1969–1971 I wrote my DPhil dissertation, “A Rhetoric of Poetic Qualities” at the University of Sussex, supervised by Prof. David Daiches. In retrospect I realized that these were my first steps in what I later called Cognitive Poetics. When I wanted to write a chapter on metrics, my supervisor said that my dissertation already exceeded the limits set by university regulations; I could write on metrics, he said, after obtaining my degree. In the years 1971–1973 I worked simultaneously on two projects, one of them being my Perception-Oriented Theory of Metre (the other—a comparative study between Mediaeval Hebrew Poetry and seventeenth-century English Metaphysical Poetry, at Southampton University) . 

My work received additional boost from my encounter with Jay Keyser. At the time I was research fellow at Southampton University, but still lived in Hove, Sussex. One day I received a phone call from a friend in London. There is here a chap, he said, who claims he has a metric theory, and is going to talk tonight at the Poetry Society. When he told me that his name was Jay Keyser, I took the first train  and went to London. At that time there was a concerted effort in the metrist community to refute the Halle-Keyser theory: its rigorous definitions threatened to rob metrists of the prerogative to say anything and be right. What could they do? They searched for counterexamples. As to myself, I found those definitions extremely useful for identifying and delineating the problems; but I offered a very different kind of solution. We had two long, fascinating discussions. 

Halle and Keyser set out to discover the rules that generate all metrical lines and no unmetrical ones. They came up with the following rule: An unmetrical line is one in which a stress maximum occurs in a weak position. A stress maximum is a stressed syllable between two unstressed syllables, as in a gárden. Two consecutive stressed syllables "neutralize" each other, so that neither of them constitutes a stress maximum, as in a róck gárden (thus, any one of them may occur in a weak position). Such an argument is circular in an important sense: a “metrical” line turned out to be one that conforms with the theory. Moreover, I had the intuition that experienced readers could perform rhythmically certain instances of “stress maximum” in a weak position, but would have had difficulties with some others. Later I obtained ample empirical confirmation of this. The concerted effort to refute the Halle-Keyser theory came up with twelve counterexamples from major English poetry. This is not very bad for a theory, and they could easily be explained (away). But I added over fifty items to the list. This didn’t mean only the increase of the sheer mass, but also yielded a corpus in which certain regularities could be detected. There are four points in the iambic pentameter line where meter could be violated (positions 3, 5, 7, 9). A random distribution would allocate 25% of violations to each of them. However, over two thirds of the instances occurred in position 7; nearly one third in position 3; in positions 5 and 9 a negligible number of rather doubtful instances can be found. I have quite good arguments to support the claim that this distribution reflects the relative difficulty of rhythmically performing a line with a stress maximum in a weak position. Furthermore, while a stress maximum in any weak position arouses a sense of frustration, return to regularity, that is, when the stress pattern and the metric pattern have again a coinciding downbeat, gives the greatest pleasure when it occurs on the last strong position; it gives reassurance of the integrity of the verse line as a whole. When I told this to Jay Keyser, he said that until I come up with a systematic theory of Rhythmical Performance, my proposal remains empty. My first reaction to this statement was shock. It took me almost half a year to recover from it. Then I decided that he was essentially right. That was Jay Keyser’s great contribution to my theory. I found no attempt to formulate such a systematic theory, only scattered attempts of instrumental analysis, sometimes within an unsatisfactory theoretical framework. So, I had to start almost from scratch, with introspection, and then learn the findings of gestalt psychology and speech research to account for my intuitions. 

Incidentally, the morning following our first meeting I ran into F. T. Prince, author of The Italian Element in Milton’s Verse, who was at the time the head of the English Department at Southampton University. Still enthusiastic, I told him about my fascinating encounter with Jay Keyser, and explained to him the stress maxima theory. To make sure he understood, he quoted from Milton’s Paradise Regained the line “And made him bow to the gods of his wives” as an instance of “unmetrical” line (gods is a stress maximum in the seventh position). When I went home, I wanted to look up the exact place of the quote in Paradise Regained; so I had to re-read the whole work with an eye on stress maxima in weak positions. To my great surprise (and satisfaction), I found no less than eighteen instances in this work only. And that was F. T. Prince’s great contribution to my theory.  

Back to home page
Back to "Occasional Papers"