In lieu of an abstract, here is a brief excerpt of the content:

Reviewed by:
  • Edgerton’s Law: The phantom evidence
  • Jared S Klein
Edgerton’s Law: The phantom evidence. By Andrew Sihler. Heidelberg: Universitätsverlag Winter, 2006. Pp. 208. ISBN 382535167X. €43.

In a series of articles appearing in Language, Franklin Edgerton (1934, 1935, 1943, 1962) formulated the expansion of Sievers’s Law that bears his name. Based most famously on evidence from Germanic (e.g. Gothic harjis ‘troop’ vs. hairdeis ‘shepherd’, nasjiþ ‘saves’ vs. sokeiþ ‘seeks’), Eduard Sievers stated (1878) that in suffixes i and iy alternated according to the structure of the preceding syllable. When it ended in a consonant preceded by a long vowel or diphthong in Proto-Germanic (hairdeis, sokeiþ) iy appeared,1 otherwise (harjis, nasjis) y.2 Sievers thought the same was true of u and uw but asserted nothing further than this. Edgerton then generalized Sievers’s Law to any position in the word, thereby rendering it anaesthetic to word-boundary, and to all the Proto-Indo-European resonants (in addition to i ∼ iy and u ∼ uw, also r ∼ ṛr, 1 ∼ ļl, n ∼ ṇn, and m ∼ ṃm), proclaiming his expanded formulation to be a phonological rule of Proto-Indo-European and part of an automatic alternation whereby for any R (= resonant), R ∼ ṚṚR stood in allophonic variation. With only minor exceptions, Edgerton based his conclusions on the evidence (i.e. scansion) of the Rigveda. His thesis was that the system he proposed for Proto-Indo-European had remained intact until shortly before the Rigveda was composed.

As widely esteemed a Sanskrit scholar and as conversant with the Vedic language as he was, when it came to the scansion of the Rigveda, Edgerton was in over his head. This is made clear in no uncertain terms by Andrew Sihler’s methodical demolition of Edgerton’s Law, which besides raises serious questions about the nature of ‘authority’ and its overreach. Edgerton’s major, indeed catastrophic, mistake was his apparent unwillingness to accept as authority in matters of Rigvedic scansion E. Vernon Arnold, whose painstaking study (1905) of the most minute details of the metrical structure of the Rigveda stands to this day as a monument representing the pinnacle of scholarship on all aspects of Rigvedic scansion.3 It is apparent that Edgerton had none but the most primitive concept of Rigvedic meter. Time after time over a span of some 150 pages S shows that Edgerton’s scansions, made in the service of his theory, are either no better than traditional scansions made without reference to any theory, decidedly worse, or effectively impossible. In the end he finds only three instances out of scores of cases that support Edgerton’s theory. Apparently, Edgerton thought it sufficient to read his CR (∼ CṚR) sequences in such a way as to get the requisite syllable count (usually eight, eleven, or twelve) within the Rigvedic line. But Rigvedic prosody is not just a matter of syllable counting. Particularly in the eleven-syllable (triṣṭubh) and twelve-syllable (jagatī) lines, the internal structure is complex, consisting of three parts (hence, the designation trimeter meter): opening (followed by caesura [= word-boundary]), break, and cadence. Trimeter lines contain either a four- or five-syllable opening followed by, respectively, a three- or two-syllable break and a cadence consisting of either four (triṣṭubh) or five (jagatī) syllables. Of these, the cadence is the most fixed stretch, typically showing trochaic (triṣṭubh: ) or iambic (jagatī: ) movement. The break is variable, but tends to show preferences for (late caesura) or (early caesura) scansion. Least constrained is the opening, which does, however, show a strong preference for [End Page 438] a long second syllable (say, or ). An idealized triṣṭubh: (jagatī) would therefore show a structure (jagatī cadence ) or (jagatī cadence: ). There are of course many variations, and Arnold compiled these. But Edgerton paid generally no heed to these internal niceties. Thus, to cite just one example among many, the received text of Rigveda 2.35.2b mántraṃ vocema kuvíd asya védat ‘We would utter a prayer. Surely he will take note of this’4 scans , a textbook triṣṭubh. But Edgerton finds two instances in the line that require adjustment by his theory...


Additional Information

Print ISSN
pp. 438-439
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.