In a recent article, Lg. 29.308 (1953), Reichardt discusses the chronology of syncopation of the connecting stem-vowel in harigasti, a name contained in the inscription on helmet B of Negau. Reichardt's argument raises an interesting combined point of method and of fact. From attested old (and dialectally unspecified) Germanic names, of which Reichardt adduces a few examples, a seeming anachronism presents itself in the occurrence of compounds with -io-as late as the 4th or 5th century, while in harigasti and in Harimellae we seem to find *-ja- reduced to -i- at a date considerably earlier than the 4th century a.d. It is, of course, possible to allege that in each case we have forms from a different dialect, and that the change occurred separately at different times in different places; but it is unsatisfying to argue from our ignorance of prehistoric Germanic dialect distributions, and Reichardt wisely refrains from introducing this answer of despair.

In contrast to other IE languages, Armenian shows a great variety of developments from IE voiceless stops; only IE *k̂ appears to yield Arm. s consistently, and will therefore be disregarded in the present discussion.

The dialect area with which this study deals is the Roman imperial administrative district of Narbonensis, the southeastern part of Roman Gaul—in essence, the modern Provence. A more exact delimitation of this region, together with a map, may be found in DAG pp. 59-62 and 14. All the extant evidence of the Gaulish language has been collected and classified by What-mough in DAG; with one or two exceptions from CIL 12, the material for this study has been drawn in its entirety from the section on Narbonensis, DAG 20-83.

In The structure of English (New York, 1952), Charles C. Fries describes a small group of intensifiers, which he labels Group D and identifies as words that pattern like very. The group is established (92) with the following frame: The (A) concert (1) may (B) not (C) be (2) very (D) good (3) then (4). Substituting in this frame, he gets, as members of Group D, very, quite, really, real, pretty, fairly, rather, awfully, awful, any, too, more, most. We are not told that these are the only members in the corpus examined by Fries; but the implication is that if the group extends beyond this list, it doesn't extend far.

The psycholinguistic approach to certain linguistic phenomena is based on two general assumptions, which most structural linguists would probably accept but would probably consider irrelevant: first, that language as a form of learned behavior is subject to the general principles which govern all learned behavior; second, that to determine what these principles are and how they function, the relative frequency of phenomena may be significant. In his description of the phonology of a language, the structural linguist has usually limited his analysis to a presentation of those sequences of phonemes which occur in a language, without noting the relative frequencies of these combinations. It is true that several studies have offered systematic statements of distribution and clustering, pointing out for example that in English a phone of the /ŋ/ phoneme does not appear in word-initial position, or that /str-/ is an initial cluster but /sθr-/ is not. But the greater-than-chance occurrence of some sequences and the less-than-chance occurrence of others has usually gone unmentioned. The contribution of a psycholinguistic analysis is to suggest that these deviations from chance are not random, but are governed by some ‘lawful’ principle.

In his article on Huichol phonemes, McIntosh notes a stylistic relationship between r and l, which he regards as phonemically separate: ‘The voiced alveolar lateral 1 is a specialized symbolic phoneme, stylistically alter-[nat]ing with r in words of endearment or diminutives. ... The alveolar flap r is varied by children to the alveolar lateral 1 and is varied to the same phoneme by adults in songs and when they talk baby talk’ (fnn. 3, 4). A similar stylistic relationship is noted between z and s, which he considers allophones of one phoneme: ‘z is varied to a nonretroflex voiceless variant by children and by adults talking baby talk and also in songs’ (fn. 2). Further examination of the language indicates that a similar stylistic alternation occurs for all alveolar phonemes. The purpose of the present paper is to interpret this alternation in terms of the concept that language is systematic; specifically, to assign the stylistic features to a definite place in the structure of Huichol. The data upon which the interpretation is based will be given in detail, and several alternative treatments will be discussed, before describing the preferred interpretation.

The relation between linguistics and logic has been discussed in a recent paper by Bar-Hillel, where it is argued that a disregard for work in logical syntax and semantics has caused linguists to limit themselves too narrowly in their inquiries, and to fall into several errors. In particular, Bar-Hillel asserts, they have attempted to derive relations of synonymy and so-called ‘rules of transformation’, such as the active-passive relation, from distributional studies alone, and they have hesitated to rely on considerations of meaning in linguistic analysis. No one can quarrel with the suggestion that linguists interest themselves in meaning or transformation rules, but the relevance of logical syntax and semantics (at least as we now know them) to this study is very dubious. I think that a closer investigation of the assumptions and concerns of logical syntax and semantics will show that the hope of applying the results which have been achieved in these fields to the solution of linguistic problems is illusory.