Fundamental concepts in phonology: Sameness and difference (review)

As reflected in the title, Fundamental concepts in phonology (FCP) explores the concepts of sameness and difference in the field of phonology. Since phonology deals with the interface between the abstract system of the native speaker’s knowledge of sounds and the physical manifestation of those sounds, the linguistic classification of those physical entities needs to be guided by a clear set of criteria for deciding what constitutes the same sound and what does not. Standard linguistic theory has asumed that the criteria for classification can be found in a segmented version of the phonetic continuum of spoken language. FCP questions the basic assumptions prevalent in many mainstream phonological analyses, such as segmentation, abstractness, monosystemicity, and derivation.

It is unarguably the case that if x is considered the ‘same as’ y in certain respects, then x and y belong to the same category, but if x and y are ‘different’, then they belong to different categories. Lodge is interested in the fundamental question of what criteria one can and should use to determine sameness and difference. He does this by investigating fields outside of linguistics (Ch. 1), within phonology (Chs. 2–6), and with respect to dialects of a single language (Ch. 7). In the following paragraphs I summarize some of the highlights of Chs. 1–6.

Ch. 1 discusses the notions of sameness and difference in areas outside of phonology, namely chemistry, biology, music, and visual representation. An analogy is drawn here between phonology and the field of chemistry: the relationship between a phoneme and its allophones can be thought of as being similar to the chemical formula H₂O and the physical realizations of a liquid, vapor, or ice, each of which occur under certain conditions. While L likes the analogy, he notes that it breaks down because humans can detect the difference between water, vapor, and ice, while the allophones of a phoneme are not normally perceived by native speakers.

Ch. 2 explores the notions of sameness and difference in phonology with data drawn from English and Scots Gaelic. The problem is illustrated with examples from English: native speakers see the regular plural allomorphs [s] and [z] in words like cats and dogs as the ‘same thing’, but they have no difficulty in treating the [s] and [z] in words like mace and maze as different. How can these intuitions be captured in phonological structure? In English, stops I broadly transcribe as /b d g/ contrast with /p t k/, but the phonetic realizations of these sounds differ according to context: word-initially, these sounds are realized along different phonetic dimensions than the ‘same’ sounds word-internally and word-finally. L describes a typical solution in mainstream phonology that relies on the phonemes /p t k b d g/ and three binary features [voice], [spread], and [constricted], and concludes that ‘an assumption of phonetically-based phonological features forces a derivational account of the relationship between the lexical form and their realizations’ (17). This type of analysis is referred to as being ‘cumbersome’, although I found that this point could have been made more explicit. L goes on to write that the onset /p/ (e.g. spin) and the intervocalic /p/ (e.g. paper) cannot be the same thing, neither at the phonological level nor at the phonetic level. Thus, only intervocalic /p/ has some aspiration and is susceptible to lenition in some varieties. From the perspective of phonology, /p/ contrasts with /b/ only intervocalically but never after /s/. In my view a more convincing example illustrating the same points can be seen in the second English example discussed in Ch. 2, namely varieties in which [t^h] and [?] are in complementary distribution in such a way that the former occurs in the onset and the latter in the coda. Phonetic similarity between [t^h] and [?] can hardly be a criterion for grouping the two sounds together into allophones of one phoneme. Much of the chapter is spent criticizing what L dubs ‘library phonology...