One consequence of the renaissance of Sieversian metrics over the past twenty years is that Old Saxon poetic meter, to which few monographs had previously been dedicated, has now been thoroughly dissected in two mammoth and exhaustive works: first by Dietrich Hofmann in a two volume study of 1991 that brings to bear the insights of A. J. Bliss's analysis of Beowulfian metrics, and now by one of the leading researchers in Old English meter. Like Hofmann, Suzuki offers an exceptionally rich synchronic description of the metrical patterns of the Heliand and Genesis, though not within the analytic framework of Bliss (who in fact is cited just once in more than 500 pages), but on the basis of Suzuki's own 1996 study of the meter of Beowulf. Yet while Hofmann (who died in 1998) was determined to analyze Old Saxon meter on its own terms, without any preconceptions about its relation to other metrical traditions, Suzuki begins with the assumption that it derives historically from a system treated here as identical to that of Beowulf, and thus his chief aim, after describing the synchronic facts in detail, is to explain how the one system evolved into the other.

Regardless of whether one accepts this initial derivational premise, the account itself of the changes involved is acute and compelling. Studies of the Old Saxon language generally assume, contra Hofmann, reduction of stress as a way of accounting for various phonological developments, including the rise of anaptyctic vowels before post- consonantal resonants, the extensive, analogically motivated restoration of syncopated vowels, and (Suzuki proposes) the preservation of postconsonantal /j/, lost in the other West Germanic languages. The most general consequence of stress reduction for the meter was that stress came to play a smaller role, and syllable quantity a greater one. More specifically, some of the chief effects were “the obscuration of the three kinds of metrical positions (the lift, the normal drop, and the heavy drop); the neutralization of basic and increased metrical types; the reappraisal of parasitic weak derivative elements occurring before and after the first lift as full metrical positions; the increasing ambiguity and partial merge[r] of normal and hypermetric verses; and the disruption of the unity between double alliteration and lift formation including resolution” (p. 340). The greater part of the book is a detailed study of these effects, documented statistically in table after table showing the separate incidence of the various verse types in the on-verse with and without double alliteration, and in the off-verse. Thus, after a linguistic introduction the chapters take up, in turn, Sievers’ five metrical types and their subtypes as represented in the Heliand, the distribution of anacrusis (with a separate, synthesizing section mopping up such diverse topics as the distinction between normal and heavy drops, the scansion of disyllabic inflections, and the treatment of quasi-compounds), resolution and its suspension, alliterative patterns, and the structure of hypermetric verses, with a superb summary and synthesis (pp. 330–344). The volume is rounded out by two appendices, on the scansion of foreign names and on the meter of Genesis (found to be similar to, but less historically developed than, that of the Heliand), a bibliography of just four pages (as opposed to the seven pages required to list all the tables in the book) and five indices, to the scansion of the two poems, and of authors cited, subjects, and verses discussed.

Given Suzuki's aim of tracing the historical development of the meter, it may seem an outlandish appraisal, but in a peculiar way the approach adopted is uniformly ahistorical. A characteristic of Old English meter that has come into particularly clear focus in recent years is its embodiment of metrical archaisms. In part this is a consequence of poets’ knowledge of verse traditions, so that, for example, although OE wundor ‘marvel’ and sēon ‘see’ were certainly a disyllable and a mono- syllable, respectively, in the poets’ own speech, they could be used with just the reverse metrical values, which the words in fact normally had before the earliest surviving poetry was recorded. Archaisms may also be due to language change that intervenes between the date of a poem's first having been recorded and of the later copy that is usually the sole witness to the poem. Thus, for example, just as in Old Saxon, syncopated vowels could be restored on an analogical basis, and though a late scribe might write a form like mōdiga ‘brave’, the meter of most poems will require the earlier, syncopated form mōdga. The latter variety of archaism is of lesser relevance to the Heliand, of which the best manuscript witnesses were made probably less than half a century after the poem's composition, though small spelling differences in a language with considerable orthographic variability can have significant consequences, especially in regard to the metrical treatment of restored and epenthetic vowels. The former variety of archaism, however, ought to have some considerable relevance, especially as it has been shown to play an important role also in Middle English alliterative meters (see, for example, Cable 1991:85–113) and in skaldic meters (for example, Sapp 2000). A meter of this sort is thus, in a sense, a diaphane, inasmuch as in the course of scansion one must look through the surface forms and take into account a historical dimension that underlies the recorded text.

It is in this sense that Suzuki's approach, like Hofmann's, is ahistorical, since the meter is for him not a diaphane but a printed page, flat and opaque. That is, regardless of the historical reasons for the form of the text (of which reasons Suzuki is of course very well aware), scansion is always based on the recorded forms, without acknowledgment of the possibility that the poet may have intended certain older forms preserved by poetic tradition. (Indeed, Suzuki frequently speaks of the Heliand poet as personally responsible for this or that metrical innovation, without recognizing the likelihood that his meter was crafted by a poetic tradition, that is by many poets working through the years.) Some will no doubt find it implausible that metrical archaisms should have played no role in Old Saxon meter, given their role in cognate metrical traditions, and given the remarkable conservatism of those traditions. At all events, certain scansional problems do arise from this practice. For example, Suzuki argues that a disyllabic desinence such as the adjective ending -ana (acc. sg. masc.) or -aro (gen. pl.) is to be treated metrically as a unit, capable of filling a metrical position (such as final position in type C) that cannot otherwise be filled by two unstressed syllables. This analysis may at first seem functionally indistinguishable from the assumption that the poet intended older, monosyllabic forms of these disyllabic endings, but the difference becomes apparent when Suzuki is obliged to stipulate exceptions to the unitary treatment of such endings, as in the scansion of craftigana Crist (Heliand 2804a), which he assumes (pp. 146–147) to have resolution of -tiga- rather than unitary treatment of -ana. That the focus on the surface form here is excessive becomes particularly plain when it is recognized that -ana is an infrequent, analogical variant of the ending, and every other of the ten or so instances of the word craftigana is spelt with an older, monosyllabic ending. Similarly, in regard to an excrescent vowel, he rejects (p. 65) the possibility that the second vowel of mêđomhord manag (3261a, 3772a) should be ignored in scansion, even though the type is otherwise unparalleled, on the ground that mêđom- is demonstrably disyllabic in mêđomhordes (1643b). But surely some metrical variability must be recognized, as in the treatment of Old English wundor and sēon (above); and note that equivalent OE -um when it represents a syllabic resonant is never a metrical syllable, probably because it was syllabified later than the other resonants. This problem of invariability in Suzuki's analysis shows itself in a variety of ways that affect his classifications, and thus his statistics.

The statistics are affected by some other debatable assumptions, some, again, related to the issue of invariability. The treatment of Kuhn's first law is unusual, excluding pronouns from the class of Satzpartikelnand treating a word such as aftar as consistently a preposition (hence unstressed) rather than variably a preposition or adverb (pp. 255–261). Oddly, then, a class 1 word such as the last in faran folc manag (1163a) is said to be too subordinated in the drop to be affected by suspension of resolution (p. 200), which is irrelevant to unstressed words. The distribution of heavy and resolvable syllables in type C is explained on the basis of an ad hoc principle of “minimum amount of prominence” (p. 115), without reference to Sievers’ rule that a short lift may not follow a resolved one in type C, even though the rule is discussed elsewhere (p. 209). It is supposed that the second constituent of a true compound may form the final drop of a verse of type C (p. 123), but two of the three examples end in ênfald ‘simple, true’, and the cognate morpheme OE -feald is rather a derivational suffix; the remaining example is thus simply anomalous, not unlikely an error. Similarly, verses like manag mârlîc thing (1295a) are assumed to bear secondary stress on -lîc, even though the suffix is acknowledged to form a quasi-compound. (Nor should -lîc be given a long vowel; OE -līc was certainly shortened after a stressed syllable, another sign that its stress was not secondary.) There is thus again a certain rigid determination against variability in the metrical treatment of individual elements. Suzuki is surprised (p. 131) that double alliteration should be greater in the type al irminthiod (1773a) than in the nearly identical Uueros gengun tô (4102b). Yet the general rule has often been remarked (first by Krackow 1903:43–4) that in Old English, true compounds must alliterate, exceptions in Beowulf (and in most other Old English poems) being exceedingly few. Certain conclusions about what is permissible in Old English verse are predicated on the witness of verses in Beowulf that most metrists (though this is unacknowledged), to avoid assuming notable coincidences, follow Sievers in regarding as resulting from the kind of scribal modernization described above as excluded from Suzuki's ahistorical reckoning, such as Sorh is mē tō secganne (473a) and hwīlum dydon (1828b; see pp. 141, 196).

The statistics are accordingly affected by many particulars of a debatable nature, and arguments that depend on a very precise reckoning of subtypes and their distribution may not convince all. Indeterminacy thus presents a notable disincentive to the degree of detail lent Old Saxon meter by both Suzuki and Hofmann. Yet the larger patterns in Suzuki's data are plain enough, and his explanation for the general differences between the meters of Beowulf and the Heliand (regardless of whether the latter really derives from the former) are admirable and make excellent sense. The book is thus an important and useful contribution to the ongoing study of early Germanic meters, for which metrists will be grateful.