Review Article: Meter in poetry

Meter in poetry (MIP) presents a unified account of the meters used in the world's poetic traditions. According to Fabb and Halle (F&H), all poetry is made up of lines and the defining feature of metrical poetry is that it involves restrictions on line length (p. 273). The aim of the book is to provide a general framework within which to characterize the various ways in which lines are measured and patterned in the world's poetic traditions. The general outlook of MIP is that of generative linguistics. Just as a linguistic theory is a theory of grammatical well-formedness, a theory of meter is a theory of metrical well-formedness. As its title indicates, the book deals only with meter, not with versification in general; topics such as rhyme, alliteration, and stanza structure are touched upon only to the extent that they are relevant to the discussion of meter.

Works with comparable theoretical goals have appeared in the past, notably Chapter 3 of Halle and Keyser (1971), Kiparsky (1977), Hayes (1983, 1989), Prince (1989), Hanson and Kiparsky (1996), Golston (1998), and Golston and Riad (2000). These were all of article size and none of them dealt with more than a few poetic traditions. MIP's empirical coverage is incomparably more vast. Here are the main headings of the table of contents: "A theory of poetic meter" (pp. 1-43); "English strict meters" (pp. 44-66); "English loose meters" (pp. 67-93); "Southern Romance" (pp. 94-132); "French" (pp. 133-152); "Greek" (pp. 153-185); "Classical Arabic" (pp. 186-213); "Sanskrit" (pp. 214-237); "Latvian" (pp. 238-250); "Meters of the world" (pp. 251-267); "The metrical poetry of the Old Testament" (pp. 268-284).

The chapter headings listed above give only a limited idea of the range of poetic traditions covered. Chapter 4 actually deals with Spanish, Italian, Galician-Portuguese, and the Saturnian verse of Latin.¹ Other languages/traditions discussed in the book are the Judeo-Spanish poetry of the Middle Ages (pp. 208ff.), Bedouin Arabic (pp. 251ff.), Hassānīya Arabic (pp. 253ff.), Chinese (pp. 255ff.), and Vietnamese (pp. 259ff.).

The system presented in MIP was foreshadowed in earlier publications, notably Halle and Keyser (1999) and Fabb (2002). But the book is self-contained. The workings of the theory and the facts that the theory purports to explain are presented with remarkable clarity. The discussion can be followed by anyone familiar with linguistic arguments and with sequential derivations.

According to MIP a meter is a set of rules and conditions. Here is how these rules and conditions operate in order to assess the metricality of a sequence of words. Taking that sequence as an input, the rules apply sequentially to construct a metrical grid. The conditions then check the resulting scansion-that is, the composite object consisting of the linguistic string and the grid. If the scansion meets all the conditions, the input linguistic sequence is deemed to conform to the meter under consideration. Example (1) outlines the derivations for checking whether two sequences are well-formed trochaic tetrameters. The sequences are Pléasure néver is at home and Pléasure is néver at home² ((22a), p. 15, and (26), p. 19). The first is a well-formed trochaic tetrameter, but the second is not.

As can be seen from the scansions in the dashed boxes in (1), a grid is made up of "Gridlines", which are rows of asterisks. Grids are constructed from top to bottom, one Gridline at a time. First, Gridline 0 is constructed by writing one asterisk under each syllable of the input sequence. After that, a set of iterative rules applies. Each rule takes as its input the sequence of asterisks created by the preceding step in the derivation. It divides that sequence into headed groups and writes an asterisk beneath the head of each group, thereby creating a new Gridline.³ In (1), for instance, here...