This book is the first in a series of four titles (see also Zurinskas Reference Zurinskas2006a, Reference Zurinskasb, Reference Zurinskasc) in which the author outlines Truespel, a ‘new pronunciation guide spelling designed as a replacement for those used in dictionaries and translation guides’ (p. 3). Zurinskas argues that Truespel improves on other phonological transcription systems because it ‘makes it possible for the first time to spell the sounds of English (and other languages) in regular letters of the keyboard in normal case’ (p. ix). He also claims, somewhat ambitiously, that Truespel can be ‘the international medium of phonetic spelling for all languages’ (p. 106), thus presumably replacing the International Phonetic Alphabet itself.
The book is divided into seventeen sections and has seven appendices. The relationship between the sections is not always clear, because the overarching structure of the book is not set out explicitly. Nevertheless, the sections appear to try to do one of two things, either (i) explain the rationale for the Truespel enterprise and set out the sound–letter correspondences used (sections 1–3, 11 and 12, although in the book the sections are not numbered, and appendix 1) or (ii) demonstrate how Truespel can be used in the analysis of a range of corpora (sections 4–10 and 13–16). In this review, I consider (i) and (ii) in turn.
The book begins by outlining the rationale for the Truespel enterprise. Zurinskas offers what he calls one of the ‘key facts’ about reading English, which is that ‘English “causes” twice as much dyslexia as a language with phonetically consistent spelling such as Italian’ (p. 5). This, he argues, suggests that a simplified ‘easy-to-use phonetic guide’ (p. 5) would be beneficial to children learning to read and spell English. Readers are told that current phonetic spellings in dictionaries are problematic because they are ‘not standardised, especially between countries’ (p. ix) and because the symbols are ‘hard-to-remember and hard-to-use’ (p. 5). Thus, the Truespel system, a set of sound–letter correspondences which uses only the letters of the English alphabet ‘fills a gap that needed to be filled’ by augmenting traditional spelling ‘in the easiest way possible through phonetic awareness’ (p. ix).
Although I agree with the sentiment of some of these points (my students, for example, would testify that the IPA is quite difficult to master), Truespel's rationale is weakened because the evidence used in support of the main arguments is misrepresented and the complexity of existing transcription systems is exaggerated. For example, the assertion that English ‘causes’ dyslexia is based on a BBC News website which reports the findings of Paulescu et al. (Reference Paulescu, Démonet, Fazio, McCrory, Chanoine, Brunswick, Cappa, Cossu, Habib, Frith and Frith2001). Zurinskas does not provide the full reference to this work, so it is unclear whether the primary source was consulted. Paulescu et al. do not argue that English ‘causes’ dyslexia, but conclude that ‘a phonological processing deficit is a universal problem in dyslexia and causes literacy problems in both shallow and deep orthographies’ (2001: 2167). Thus, they say, the problem in dyslexia is phonological processing, not the spelling system per se. Truespel is unlikely to help here. Zurinskas is right that the IPA is more complex than Truespel, but this is inevitable because whilst the IPA is able to represent all sounds of human language, Truespel is based on just one variety of English (with an additional 20 sound–spelling correspondences for all other languages). Another criticism of the IPA is that the symbols are not ‘keyboard friendly’ (p. ix). This is not necessarily true, given the increasing use of Unicode. As well as this, the SAMPA (Speech Assessment Methods Phonetic Alphabet) system of transcription, which has been around since the late 1980s (see Wells Reference Wells, Gibbon, Moore and Winski1997), is entirely keyboard-accessible. Zurinskas briefly compares Truespel to SAMPA, concluding that the latter ‘is not intuitive to English speakers . . . also, it uses special characters (upper case, quote mark for emphasis, number 3 etc). This is not good. Better would be to do as Truespel does to make phonetic respelling most user friendly to English, which is the most important language’ (p. 87). I am not convinced of the value of making a transcription system ‘English-friendly’, particularly if the aim is for Truespel to be used as ‘the international medium of phonetic spelling for all languages’ (p. 106). Furthermore, assertions that Truespel is ‘better’ than SAMPA from a user perspective are unsubstantiated because no direct comparisons are carried out.
Even if the size of the ‘gap that needed to be filled’ (p. ix) is exaggerated, it is clear that Zurinskas has thought quite hard, on the whole, about the sound–letter correspondences required for USA English. Space constraints restrict full discussion of each of the correspondences here. Instead, two illustrations are provided below, namely (1), the sound–letter correspondences provided for USA English vowels, and (2) a translation from standard English to Truespel transcription.
(1) Truespel vowels (p. 108)
(2) Translation from standard English to Truespel (p. 14)
Standard English
That quick beige fox jumped in the air over each thin dog. Look out I shout, for he's foiled you again.
Truespel
That kwik baezh faaks jumpd in thee air oever eech thhin daug. Look out ie shout, for heez foild yue uggen.
The sound–spelling correspondences in Truespel show flashes of phonetic awareness, but the system has been disadvantaged by a lack of proper engagement with phonetic and phonological facts. For example, in the transcription in (2), Truespel does well to maintain a distinction between /ð/ in the (spelled 〈th〉), and /θ/ in thin (spelled 〈thh〉), but the past tense suffix of jumped and foiled is represented identically (with 〈d〉), despite the phonetic difference. A further problem is that because Truespel uses terms such as ‘long vowel’, ‘short vowel’ and ‘combined’ in a non-standard (and sometimes unfathomable) way, it lacks the ability to distinguish between long and short monophthongs and diphthongs. The representation of weak vowels is another concern. Schwa, for example, is missing from Truespel. This means the vowels in words like up, jump, the and again must be represented with the same letter (〈up〉, 〈jump〉, 〈thu〉, 〈uggen〉). This is not an oversight on Truespel's part, being justified as follows: ‘Schwa represents unstressed short vowels, but overlaps already defined sounds. By eliminating the vague use of schwa, the Truespel dictionary is far more accurate than other pronunciation guides that do use schwa’ (p. 33). The initial point here is basically correct, of course, but to remove schwa completely from the phonemic system is unhelpful, and certainly does not help make Truespel more accurate than other transcription systems. Part of the overarching problem with the Truespel system is that it is not clear whether it is meant to be used only for phonemic transcription or whether it strives for phonetic accuracy. At first, I assumed the former position, because readers are told early in the book that Truespel includes the ‘40 phonemes’ of USA English (p. ix). However, later, suggesting the latter position, we learn that Truespel includes a mechanism for transcribing the glottal (or ‘glaa'ool’) stop. This seems out of place, and brings into focus the question that if certain allophones are transcribable in Truespel, why not others? Why is there no mechanism for transcribing the alveolar tap, for example, which is so common in USA English? These questions are not addressed in the book, largely because the relationship between phonemes and their allophones is not tackled in any way.
Once the rationale for Truespel and its sound–spelling correspondences have been outlined, Zurinskas analyses three corpora. These are (i) the Truespel dictionary; (ii) a sample of the London Times newspaper, converted to Truespel; and (iii) a sample of 60 words/phrases of 14 languages, again converted to Truespel. This is a brave attempt at showing what Truespel can do, but there are a number of problems. The first relates to the lack of detail provided about each corpus. For example, although readers are told that the Truespel dictionary consists of around 60,000 words (including roots and affixes), no information is provided about how the words were selected, how the dictionary was compiled or how it works. Furthermore, readers learn that the phonological transcriptions in the dictionary are based on ‘the words of a talking dictionary voice files, with intentional non-reference to any existing phonetic guide to avoid any influence’ (p. 9), but no information is offered about the talking dictionary that is used (the Merriam–Webster website is referenced, but Zurinskas explains that this was not the pronunciation guide that provided Truespel's model). The London Times data consists of words that appeared 100 times or more in the newspaper ‘over a period of time’ (p. 23) and is problematically taken to be representative of ‘phonemes spoken in USA English’ (p. 23) because it ‘represents the frequency of commonly used words in average conversation’ (p. 9). There is no discussion of the validity of these claims, or even an awareness of the danger of comparing spoken data to written data without caution. The data for the third corpus is taken from 60 words/phrases from 14 languages, taken from the ‘Word of the Day’ section of a travel website (http://www.travlang.com). Once again, no rationale is given for the selection of words included in the sample.
The rest of the book compares these corpora in order to make observations about the frequency of phonemes in the three datasets. The London Times data is directly compared to the language sample data, for example, in section 5. A table is presented which illustrates how frequently each Truespel phoneme appears in each corpus. This is an interesting exercise in principle, but because of the different ways in which the corpora were compiled, I cannot see what this sort of direct comparison achieves. The phoneme count of the language sample data relies completely on which 60 words/phrases were included, and because the rationale for this is not described, readers cannot make comparability judgements across corpora. My understanding of the comparison was not helped by the fact that the data is provided in a single table, spanning two pages, with no additional discussion. Thus, the value of the comparing data in this way is not explicitly set out. A potentially more useful analysis, in section 9, presents an illustration of the ways in which USA English phonemes are spelled. However, because this analysis consists of a series of tables spanning 40 pages with little intervening discussion, the potential has to be realised on the part of the reader, rather than being explicitly foregrounded by the author. Unfortunately, the same can be said of other sections of the book. Too often, readers are expected to glean relevant information from tables densely packed with information. This information would have been more accessible, and so would have allowed Zurinskas's points to shine through with greater ease, if there had been accompanying discussion and debate which engaged more readily with existing literature.
Overall, whilst the Truespel enterprise is an interesting exercise, it is not sophisticated enough to become the international medium for phonetic transcription, as Zurinskas hoped. It seems clear that although Truespel is based on a basically sound ‘ear for phonetics’, its lack of engagement with the disciplines of phonetics and phonology is a fundamental flaw. This point has not gone unnoticed in the phonetics and phonology community, with Truespel being discussed on a number of electronic mailing lists over the past few years. If phoneticians and phonologists are to be convinced that Truespel is useful, the book needed to provide a detailed, direct comparison between it and other transcription systems, but such comparison is lacking. Indeed, the IPA receives very little consideration at all. The book also needed to take greater care to be academically rigorous. Where references to external sources are provided, they are to websites rather than to academic work, and even then they are misrepresented. This lack of academic rigour means that the book is unlikely to be useful as a text in university phonetics and phonology courses. That said, I have had interesting and pedagogically beneficial discussions with my students about the points Zurinskas makes, in particular by asking them to engage with the material and make observations corroborated by phonetic and phonological theory. This was a surprisingly fruitful use of a book which otherwise falls short of achieving its aims.
