<?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en"> <head> <title>Wells: Syllabification and allophony</title> <meta http-equiv="content-type" content="text/html; charset=x-UNICODE-2-0-UTF-8" /> <meta name="description" content="Argues for a view of syllabification in English phonetics in which codas are maximized"> <meta name="keywords" content="syllable, syllabic, phonotactics, English phonetics, syllabification, syllabication, 'syllable boundary', 'syllable division', onset, coda"> <style type="text/css"> body {background-color:#eeeeee; } h1 {font-family:"times new roman", serif; font-weight:bold; font-size:x-large; color:red;} h2 {color:blue; font-family:"times new roman", serif; background-color:transparent;} h3 {color:green; font-family:"times new roman", serif; background-color:transparent;} h4 {font-style:italic; font-weight:normal; font-size: 115%; font-family:"times new roman", serif; margin-left:15%; } p, li {color:navy; background-color:transparent; margin-left:15%; font-family:"gentium","lucida sans unicode","arial unicode ms"; font-size:90%; line-height:1.2; text-align:justify; text-indent:3%;} blockquote {margin-left:25%; color:navy; background-color:transparent; text-indent:-5%; font-family:"gentium","lucida sans unicode","arial unicode ms"; font-size:90%; line-height:1.2; text-align:justify; } tr {color:navy; background-color:transparent; font-size:85%} div.indent {margin-left:10em;} div.small {font-size:70%; margin-right:30%; font-family:"lucida sans unicode","arial unicode ms", "gentium"; } div.italic {font-style:italic;} </style> <script language="JavaScript" type="text/javascript"> /* Made by Martial Boissonneault 2001 http://getElementById.com/ May be used and changed freely as long as this msg is intact Visit http://getElementById.com/ for more free scripts and tutorials */ function ChangeFontFamily(id, family) { document.getElementById(id).style.font-family = family; } </script> </head> <body> <!-- <a href="javascript://" onclick="ChangeFontFamily('lyr','Arial')">Change to Arial</a> <a href="javascript://" onclick="ChangeFontFamily('lyr','gentium')">Change to Gentium</a> --> <h1>Syllabification and allophony</h1> <h2>J.C. Wells<br>University College London<br>England</h2> <p><div class="small">This document uses Unicode to encode IPA phonetic symbols. If you cannot see a schwa here [&#601;] on your screen, users of Windows 95/98/ME/NT/2000/XP and Macintosh OSX please <a href="lsansuni.ttf">download Lucida Sans Unicode</a> free of charge and install it on your system, and/or install a newer browser.</div> <br /> <div class="small">This article was published in Susan Ramsaran (ed.), <i><b>Studies in the pronunciation of English</b></i>, A commemorative volume in honour of A.C. Gimson (London and New York: Routledge, 1990), pages 76-86. </div> <br /> <div class="small">It argues for a view of syllabification in English phonetics in which not only onsets but also codas are maximized in stressed syllables. This is the hypothesis that underlies the syllable divisions shown in my <i>Longman Pronunciation Dictionary</i>.</div> </p> <h3>Introduction</h3> <p>In his complete revision of Daniel Jones&#x2019;s <i>English Pronouncing Dictionary</i> (1977), Gimson did away with the hyphen symbol which Jones had considered &#x2018;necessary ... [as] a means of showing &#x201C;syllable separation&#x201D; ... in all circumstances where the absence of suitable marking might lead to ambiguity in the interpetation of a phonetically transcribed word&#x2019; (Jones 1963: xxvi). Thus in the twelfth edition, his last, Jones transcribed <i>toe-strap</i> as /&#712;t&#601;&#650;-str&#230;p/ but <i>toast-rack</i> as /&#712;t&#601;&#650;st-r&#230;k/. In Gimson&#x2019;s revision these words appear simply as /&#712;t&#601;&#650;str&#230;p, &#712;t&#601;&#650;str&#230;k/. While recognising that &#x2018;the situation of the syllable division (juncture) has implications for the duration and quality of the sounds involved&#x2019;, Gimson justifies his decision on the grounds that &#x2018;such divisions and their implications for pronunciation are generally evident from the orthography and from the meaningful segmentation (morpheme boundaries) of the word&#x2019; (Jones 1977: xiv).</p> <p>I think this decision was unfortunate. Not only did it mean the removal of information which is undoubtedly part of the specification of a word&#x2019;s pronunciation, but it also made it impossible to show competing pronunciations that differ only in syllabification. Jones was able to indicate that the word <i>teaspoon</i> is commonly pronounced in RP as if it were monomorphemic: he gave /&#712;ti&#720;spu&#720;n/ as the first pronunciation, /&#712;ti&#720;-spu&#720;n/ only as a less common variant. Under Gimson&#x2019;s revision this information disappeared from the dictionary. Another of Jones&#x2019;s examples is <i>shellfish</i> /&#712;&#643;el-fi&#643;/, where the /l/ &#x2018;is treated as if it were final and is consequently pronounced long&#x2019;, while the absence of any hyphen in <i>selfish</i> /&#712;selfi&#643;/ implies that the /l/ &#x2018;is not so treated, but is short&#x2019; on account of the presence of the following /f/ (1963: xxvii). In Gimson&#x2019;s revision these two words are transcribed as if they were perfect rhymes (which they are not): /&#712;&#643;elf&#618;&#643;, &#712;self&#618;&#643;/.</p> <p>Both <i>shellfish</i> and <i>selfish</i> contain internal morpheme boundaries. Yet, as Jones implicitly recognised, it is not necessary to give any explicit notational recognition to the boundary between <i>self-</i> and <i>-ish</i> in <i>selfish</i>. If we compare <i>selfish</i> with the monomorphemic <i>dolphin</i> /&#712;d&#594;lf&#618;n/, we find that they are rhythmically identical. In both cases it is clear that the /f/ exerts its duration-reducing influence on the preceding sonorant sequence /&#594;l/, an influence which is blocked in the case of <i>shellfish</i>. Thus the only morpheme boundaries that for Jones need to be reflected in the phonetic transcription are those that upset the implicitly expected syllabification.</p> <p>Many analysts would claim that the syllable boundary in <i>dolphin</i> lies between the /l/ and the /f/. I believe this to be wrong. Rather, the correct syllabification is /&#712;d&#594;lf.&#618;n/, as an adequate account of English syllabification ought to predict. And this syllabic division is not &#x2018;evident from the orthography and from the [morphology]&#x2019;. (Here and henceforth I symbolise all word-internal syllable boundaries explicitly, using a dot /./ for the purpose.)</p> <p>In his <i>Introduction</i>, in the discussion justifying the taking of the word as the basis for phonemic analysis, Gimson (1980: 55) refers to the two phrases <i>plum pie</i> and <i>plump eye</i>. If we were to ignore the word boundaries they contain, he argues, we should be faced with the necessity to recognise &#x2018;two /p/ phonemes, one with, one without aspiration&#x2019;. As long as we recognise word boundaries in our phonemic analysis, this absurdity is avoided. In his section on juncture (1980: 255-6) he rightly modifies the argument to allow for the possible influence of morpheme boundaries, too,as <i>highness</i> vs. <i>high-ness</i>, <i>nitrate</i> vs. <i>night-rate</i>.</p> <p>For Gimson the syllable is relevant mainly as a possible phonotactic category (1980: 56) or as a category to which phonotactic constraints may be referred (1980: 57-8). He is sceptical of the first, and for the second prefers the word (1980: 237-53). Yet English has a fair number of important allophonic rules which can best be described by specifying &#x2018;syllable boundary&#x2019; as part of the conditioning environment. It is this fact which makes syllabification phonologically relevant.</p> <h3>Phonetic rules conditioned by syllable boundary</h3> <p>Prime among such rules are the one which conditions presence or absence of aspiration, and the one which determines the duration of vowels and sonorant consonants before a fortis consonant. Others include those specifying allophones of /t/ and of /r/, those determining possible epenthesis and elision, and those governing many minor duration details. Most were pointed out by Jones (1931, 1956).</p> <h4>Aspiration</h4> <p>English /p t k/ are aspirated when initial in a full-vowelled syllable. Elsewhere they have less aspiration or none. Hence the aspiration of the /p/ after the /m/ in <i>plum pie</i>, but its absence in <i>plump eye</i>. The /t/ of <i>a tease</i> is aspirated, as is that of <i>attack</i> /&#601;.&#712;t&#230;k/, but not that of <i>at ease</i>.</p> <h4>Pre-fortis clipping</h4> <p>This is the name which some of us have come to adopt for the rule making the /el/ of <i>shelf</i> durationally different from the /el/ of <i>shelve</i>, and the /i&#720;/ of <i>feet</i> different from that of <i>feed</i>. (Gimson referes sometimes to &#x2018;shortness&#x2019; of the sounds involved, sometimes to &#x2018;reduction&#x2019;. Calling such sounds &#x2018;short&#x2019; leads to confusion when pairs of phonemically distinct vowels such as /i&#720;/ and /&#618;/ are also categorised as &#x2018;long&#x2019; and &#x2018;short&#x2019; respectively; calling them &#x2018;reduced&#x2019; is to be avoided since this term for most phoneticians denotes change of quality, a &#x2018;reduced&#x2019; vowel being of the [&#601;] type. The term &#x2018;clipping&#x2019; avoids these difficulties.)</p> <p>English vowels are subject to pre-fortis clipping, then, when they are followed by a fortis consonant within the same syllable. The /f/&#x2019;s in <i>self, selfish</i> /&#712;self.&#618;&#643;/, and <i>dolphin</i> /&#712;d&#594;lf.&#618;n/ trigger clipping, but not those in <i>shellfish</i> /&#712;&#643;el.f&#618;&#643;/ or <i>funfair</i> /&#712;f&#652;n.fe&#601;/. So do the /t/ in <i>feet</i> and the /&#679;/ in <i>feature</i>, but not the /p/ in <i>fee-paying</i> or the /k/ in <i>tea-kettle</i>. The vowel /&#230;/ undergoes pre-fortis clipping in <i>lap, lamp, happy</i> /&#712;h&#230;p.&#618;/, and <i>hamper</i> /&#712;h&#230;mp.&#601;/, but not in <i>slab</i> or <i>clamber</i>.</p> <h4>/t/ tapping</h4> <p>&#x2018;For some speakers [of RP], and generally in American English, /t/ is realized in weakly accented intervocalic positions as a lenis, rapid tap resembling a /d/ or one tap [&#638;], e.g. <i>butter, latter, put it over there</i>&#x2019; (Gimson 1980: 164-5). More generally expressed, those who tap /t/ do so when it is syllable-final. (Other constraints are that the /t/ is not preceded by an obstruent, and is immediately followed by a vowel in the next syllable). Candidates for tapping include the /t/ of <i>might I</i>, but not that of <i>my tie</i>.</p> <h4>/t/ glottalling</h4> <p>&#x2018;Increasingly, /t/ in syllable final positions is reinforced or replaced by a glottal closure unless a vowel or syllabic /n/ or /l/ follows&#x2019; (Gimson 1980: 165). Rightly said: though nowadays a following syllabic /n/ exhibits a distinctly waning influence in this regard, and the other constraints are weakening, too. (But a preceding obstruent, as in <i>beastly</i>, continues to block glottalling.) A glottal stop is a distinct possibility for the syllable-final /t/&#x2019;s of <i>Atkins, Gatwick,</i> and <i>jointly</i>, but not for the syllable-initial /t/&#x2019;s of <i>twice</i> or <i>atomic</i>.</p> <h4>/r/ allophony</h4> <p>&#x2018;Within RP, the frictionless continuant variety [&#633;] is frequently replaced by an alveolar tap [&#638;] in intervocalic positions, e.g. <i>very, sorry, ... for ever ...</i>&#x2019; (Gimson 1980: 207). Those whose speech follows this pattern may also have [&#638;] in <i>your ice</i>, but not in <i>your rice</i>. That is to say, [&#638;] is a realisation of their /r/ in syllable-final position but not in syllable-initial. Others (I am surely not alone in this) have markedly greater lip action (protrusion, rounding) in syllable-initial /r/ than in syllable-final: greater labialisation in <i>red</i> /red/, <i>key-ring</i> /&#712;ki&#720;.r&#618;&#331;], <i>her rice</i> /h&#604;&#720; &#712;ra&#618;s/ than in <i>berry</i> /&#712;ber.&#618;/, <i>fearing</i> /&#712;f&#618;&#601;r.&#618;&#331;/, <i>her ice</i> /h&#604;&#720;r &#712;a&#618;s/. (There are also speakers who tap and/or labialise independently of syllable position.)</p> <h4>Plosive epenthesis</h4> <p>&#x2018;Few RP speakers regularly maintain the distinction between /ns/ and /nts/ ..., /nts/ tending to be used in all cases&#x2019; (Gimson 1980: 187). Yet although many people pronounce [t] in <i>fence</i> and <i>dance</i>, no-one does in <i>inside</i> or <i>rain-soaked</i>. Epenthesis happens only within a syllable, not across a syllable boundary.</p> <h4>Elision of /t/ and /d/</h4> <p>&#x2018;In PresE simplification of clusters continues to take place, especially involving the loss of the alveolars /t, d/ when medial in a cluster of three consonants&#x2019; (Gimson 1980: 236). The /t/&#x2019;s in <i>strong</i> and <i>mistrial</i> prima facie conform to Gimson&#x2019;s conditions for elision. Yet no-one elides them. Elision of English /t/ and /d/ is a possibility only when /t/ or /d/ is part of a syllable-final cluster. Thus in <i>first-rate</i> /&#716;f&#604;&#720;st.&#712;re&#618;t/ the word-medial /t/ is elidable; in <i>mistrial</i> /&#716;m&#618;s.&#712;tra&#618;&#601;l/ it is not.</p> <h4>Other duration rules</h4> <p>The difference between <i>a name</i> and <i>an aim</i> is well-known (Gimson 1980: 295). It is easily accounted for by distinguishing between the longer, stronger allophone in syllable-initial position (<i>annoy</i> /&#601;.&#712;n&#596;&#618;/, <i>a name</i>) and the shorter, weaker, perhaps tapped allophone in syllable-final pre-vocalic position (<i>penny, an aim</i>). More subtly, /n/ is shorter in syllable-final clusters (<i>standing</i> /&#712;st&#230;nd.&#618;&#331;/, <i>brandish</i> &#x2018;wield, wave&#x2019; /&#712;br&#230;nd.&#618;&#643;/) than when a following consonant is in a separate syllable (<i>bran-dish</i> &#x2018;dish for bran&#x2019; /&#712;br&#230;n.d&#618;&#643;/).</p> <h3>The main syllabification principle</h3> <p>If allophonic rules are to be allowed to refer to syllable boundaries as part of their conditioning environments, we need a principled way of specifying the location of such boundaries. I propose that English syllabification is governed by a straightforward principle:</p> <blockquote> (1) Subject to certain conditions (discussed below), consonants are syllabified with the more strongly stressed of two flanking syllables. </blockquote> <p>Thus the /k/ in <i>packet</i> belongs to the first, stressed, syllable. This analysis is supported by its homophony with <i>pack it</i>: both are /&#712;p&#230;k.&#618;t/. The /f/ of <i>dolphin</i> belongs in the first syllable: /&#712;d&#594;lf.&#618;n/ has the same rhythm as <i>selfish</i> /&#712;self.&#618;&#643;/, where this division is supported by the morphology. The /p/ in <i>happy</i> belongs in the first syllable, as evidenced by its relative lack of aspiration and by the pre-fortis clipping of the /&#230;/: /&#712;h&#230;p.&#618;/. Both the /n/ and the /t/ of <i>enter</i> belong in the first syllable, since the /t/ triggers clipping of both the /e/ and the /n/. The /p/ of <i>typing</i> /&#712;ta&#618;p.&#618;&#331;/ conditions clipping of its syllable-mate /a&#618;/: compare <i>tiepin</i>, where the /p/ exerts no such influence. (Such clipping of the /a&#618;/ as there is in this latter word falls under the different heading of &#x2018;rhythmic clipping&#x2019;, the isochronising effect of unstressed syllables on a preceding stressed syllable.)</p> <p>Similarly, <i>crisis</i> is /&#712;kra&#618;s.&#618;s/: compare <i>rising</i> /&#712;ra&#618;z.&#618;&#331;/, with a lenis syllable-final consonant, hence less clipping. The rhythmic difference between <i>hearty</i> /&#712;h&#593;&#720;t.&#618;/ and <i>hardy</i> /&#712;h&#593;&#720;d.&#618;/ has the same explanation, and is to be referred to the durational difference between <i>heart</i> and <i>hard</i>. In <i>driver</i> /&#712;dra&#618;v.&#601;/, as in thousands of other words, the phonology parallels the morphology (<i>pace</i> Fudge 1969: 20). In <i>banker</i> we see this even more clearly (pre-fortis clipping, /&#712;b&#230;&#331;k.&#601;/); <i>anchor</i> rhymes with it perfectly, but <i>fan club</i> has a different rhythm.</p> <p>As the influence exerted by suffixes causes the stress to shift, so the syllabic affiliations of consonants change. In <i>note</i> and <i>noting</i> /&#712;n&#601;&#650;t.&#618;&#331;/ the /t/ of <i>not(e)</i> is syllable-final, but in <i>notation</i> /n&#601;&#650;.&#712;te&#618;&#643;.n/ and <i>annotate</i> /&#712;&#230;n.&#601;.te&#618;t/ it is syllable-initial and aspirated. In <i>attest</i> /&#601;.&#712;test/ the first /t/ is strongly aspirated, attracted into the second syllable by the stress; in <i>attestation</i> /&#716;&#230;t.e.&#712;ste&#618;&#643;.n/ it has less aspiration or none, since the second syllable is now unstressed while the first has secondary pre-tonic stress, which makes it capture the /t/ back. In <i>apply</i> /&#601;&#712;pla&#618;/ the /l/ is voiceless, as it carries the aspiration of the syllable-initial /p/; in <i>application</i> /&#716;&#230;p.l&#618;.&#712;ke&#618;&#643;.n/ it is less so. In <i>magnetic</i> /m&#230;g.&#712;net.&#618;k/ the /t/ is syllable-final and a candidate for possible tapping; in <i>magnetism</i> /&#712;m&#230;g.n&#601;.&#x2080;t&#618;z.&#601;m/ the tertiary (post-tonic) stress on /&#618;z/ is sufficient to attract the /t/ into syllable-initial position, triggering aspiration while blocking tapping.</p> <h4>Stress levels</h4> <p>The expression &#x2018;more strongly stressed&#x2019; in (1) has to be interpreted as referring to position on a five-point scale:</p> <ol><li>primary word stress; <li>pre-tonic secondary stress; <li>tertiary (post-tonic) stress; <li>unstressed but with full vowel; <li>weak (reduced) vowel.</ol> <p>All five grades are illustrated in the word <i>substitution-product</i> /&#716;s&#652;b.st&#618;.&#712;tju&#720;&#643;.&#601;n.&#x2080;pr&#594;d.&#652;kt/, where the syllables are of grades 2, 5, 1, 5, 3, 4 respectively. In <i>magnitude</i> the /t/ goes with the final syllable, /&#712;m&#230;g.n&#618;.tju&#720;d/, because the third syllable (grade 4) outranks the second (grade 5).</p> <h4>Adjacent syllables of equal rank</h4> <p>The only cases in English where immediately adjacent syllables have equal grade are those involving weak vowels (grade 5). They are governed by the principle:</p> <blockquote> (2) Where adjacent syllables are of equal grade, consonants are (again subject to stated conditions) syllabified with the leftward syllable. </blockquote> <p>The /t/ allophones in <i>carpeting</i> /&#712;k&#593;&#720;p.&#618;t.&#618;&#331;/, <i>covetous</i> /&#712;k&#652;v.&#618;t.&#601;s/ and <i>purity</i> /&#712;pj&#650;&#601;r.&#601;t.&#618;/ make this clear.</p> <p>Americans seem agreed that they can tap the /t/ in <i>quality</i>, but not the /t/ in <i>politics</i>. This must be because the /&#618;/ of <i>-ics</i> counts as a full vowel, sufficient to outrank the weak vowel of the second syllable and thus capture the /t/; but the /&#618;/ (or /i&#720;/) at the end of <i>quality</i> counts as weak, leaving the /t/ syllable-final in the second syllable, by (2), and thus tappable.</p> <h4>Words like <b>apex</b></h4> <p>According to (1), the /p/ in <i>apex</i> should go with the first, stressed, syllable: /&#712;e&#618;p.eks/. This seems correct for RP, though it may well not be correct for some other varieties of English. If we consider the contrived examples <i>A pecks</i> (in some classification of kinds of <i>peck</i>, with <i>A pecks, B pecks, C pecks ...</i>) and a possible brand name <i>ape-x</i> (compare <i>Timex, Durex</i>), it is clear that their syllabification follows the morphology: <i>A pecks</i> /&#712;e&#618;.peks/, <i>ape-x</i> /&#712;e&#618;p.eks/. And the ordinary word <i>apex</i> is like the second and unlike the first. The /p/ of <i>apex</i> does indeed condition pre-fortis clipping of the /e&#618;/, and must therefore be in the first syllable.</p> <h3>The morpheme boundary condition</h3> <p>Morpheme boundaries such as those between the elements of a compound normally block the operation of (1). The /p/ of <i>fee-paying</i> remains initial in the second syllable, so that there is no pre-fortis clipping of the /i&#720;/ (compare <i>deep</i>). The same applies in <i>re#print</i> (n.) /&#712;ri&#720;.pr&#618;nt/ (compare <i>reaper</i> /&#712;ri&#720;p.&#601;/) and <i>pre#suppose</i> /&#716;pri&#720;.s&#601;.&#712;p&#601;&#650;z/ (compare <i>priest</i>). There is pre-fortis clipping of the /a&#618;/ in <i>hyphen</i> /&#712;ha&#618;f.&#601;n/, but not of that in <i>high-faluting</i> /&#716;ha&#618;.f&#601;.&#712;lu&#720;t.&#618;&#331;/. We need the following as a condition on the main principle:</p> <blockquote> (3) In polymorphemic words, consonants belong to the syllable appropriate to the morpheme of which they form a part. This applies only to synchronic, psychologically real morphemes. </blockquote> <p>It is this condition which explains the potential rhythmic differences between <i>Roman</i> /&#712;r&#601;&#650;m.&#601;n/ and <i>bow#man</i> /&#712;b&#601;&#650;.m&#601;n/, <i>bonus</i> /&#712;b&#601;&#650;n.&#601;s/ and <i>slow#ness</i> /&#712;sl&#601;&#650;.n&#601;s/, or <i>highness</i> (regal term of address) /&#712;ha&#618;n.&#601;s/ and <i>high-ness</i> (quality of being high) /&#712;ha&#618;.n&#601;s/ (Sharp 1960). A recent example from the world of popular music is <i>prima donna</i> /&#716;pri&#720;m.&#601;.&#712;d&#594;n.&#601;/ vs. <i>pre-Madonna</i> /&#716;pri&#720;.m&#601;.&#712;d&#594;n.&#601;/.</p> <p>For the many English suffixes which begin with a vowel, (3) is irrelevant. By (1), words such as <i>bigger, oldest, putting, horses, zealous, scenic</i> already receive syllable boundaries coinciding with the morpheme boundaries: /&#712;b&#618;g.&#601;, &#712;&#601;&#650;ld.&#618;st, &#712;p&#650;t.&#618;&#331;, &#712;h&#596;&#720;s.&#618;z, &#712;zel.&#601;s, &#712;si&#720;n.&#618;k/.</p> <p>Certain suffixes do not count as &#x2018;psychologically&#x2019; real under (3). An example is the <i>&#x2011;ful</i> of <i>aweful, careful</i>, words which are pronounced with pre-fortis clipping of the stressed vowel, which vowel must therefore have captured the /f/ from the suffix. (Compare <i>awe-ful</i> &#x2018;full of awe&#x2019;, where the /&#596;&#720;/ is unclipped.) In proper names, <i>&#x2011;ton</i> and <i>&#x2011;son</i> are usually treated as if not separate morphemes, as is evidenced by the pre-fortis clipping usual in <i>Barton</i> /&#712;b&#593;&#720;t.n/ and <i>Dawson</i> /&#712;d&#596;&#720;s.n/ and by the possible epenthesis in <i>Benson</i> /&#712;ben(t)s.n/. Yet <i>&#x2011;ford</i>, I think, does behave phonetically as a separate morpheme: <i>Crayford</i> /&#712;kre&#618;.f&#601;d/. The morpheme boundary before <i>&#x2011;ism</i> does not inhibit capture of the /t/ in <i>magnetism</i>, as we saw above; but in less familiar words the syllabification tends to follow the morphology: <i>Bonapartism</i> /&#712;b&#601;&#650;n.&#601;.p&#593;&#720;t.&#716;&#618;z.&#601;m/, <i>puppetism</i> /&#712;p&#652;p.&#618;t.&#716;&#618;z.&#601;m/. It is again American tapping that shows clearly that the morpheme boundary before the suffix <i>&#x2011;ise</i> does not inhibit /t/-capture in <i>magnetise</i> /&#712;m&#230;g.n&#601;.ta&#618;z/, <i>sensitise, sonnetise</i>. Introspecting, I find that I treat <i>&#x2011;dom</i> inconsistently, saying <i>freedom</i> /&#712;fri&#720;d.&#601;m/ but <i>boredom</i> /&#712;b&#596;&#720;.d&#601;m/.</p> <p>Some speakers, aware of etymology and meaning, may have an unclipped /a&#618;/ in <i>tri#pod</i>; but not, presumably, in <i>tripos</i> /&#712;tra&#618;p.&#594;s/. In general, this whole area of presence/absence of phonetic correlates of morpheme boundaries is still far from fully explored.</p> <h3>The phonotactic condition</h3> <p>The main syllabification principle does not operate in such a way as to lead to consonant clusters which are phonotactically ill-formed. Thus (1) is subject to the condition:</p> <blockquote> (4) Phonotactic constraints on syllable structure (as established with reference to monosyllables) are not violated. </blockquote> <p>This means, for example, that <i>timber</i> is syllabified as /&#712;t&#618;m.b&#601;/, since /mb/ is not a possible final cluster: /b/ cannot be captured into the stressed syllable. Similarly, <i>anger</i> is /&#712;&#230;&#331;.g&#601;/, at least in RP. But <i>tender</i> is /&#712;tend.&#601;/, /nd/ being a permitted cluster (<i>stand</i>). Notice how neatly this fits with permitted initial /Cl/ clusters: <i>tumbler</i> /&#712;t&#652;m.bl&#601;/, <i>English</i> /&#712;&#618;&#331;.gl&#618;&#643;/, but <i>chandler</i> /&#712;t&#643;&#593;&#720;nd.l&#601;/ (just as we have /bl&#x2011;, gl&#x2011;/, but no /dl&#x2011;/).</p> <p>Questions arise over /r/ and /&#658;/. Although final /r/ does not occur in RP in words pronounced in isolation, it does occur in connected speech, and in such a way as to make clear that /r/ can be syllable-final (see the discussion of /r/ allophony above). Linking /r/, both internal and external, is indeed syllabified with the preceding vowel. Accordingly we need not hesitate to analyse <i>bleary, sharing</i> as /&#712;bl&#618;&#601;r.&#618;, &#712;&#643;e&#601;r.&#618;&#331;/ and hence the phonetically comparable <i>weary, Mary</i> as /&#712;w&#618;&#601;r.&#618;, &#712;me&#601;r.&#618;/. Equally, <i>sorry</i> is /&#712;s&#594;r.&#618;/ and <i>spirit</i> /&#712;sp&#618;r.&#618;t/. Final /&#658;/ is admittedly restricted to loan-words (<i>rouge, beige</i>), but this is sufficient justification for us to accept the analysis <i>measure</i> /&#712;me&#658;.&#601;/.</p> <p>Notice that (1) is already sufficient to reflect the phonotactic constraint that disallows the occurrence of short vowels finally in a stressed syllable. <i>Better</i> is /&#712;bet.&#601;/ rather than */&#712;be.t&#601;/ not only because the /t/ triggers pre-fortis clipping and is tappable, but also because a syllable /&#712;be/ would not be phonotactically well-formed.</p> <p>A difficulty arises in the case of words like <i>nostalgic, posterior, fastidious</i>. Is the correct analysis /n&#594;.&#712;st&#230;ld&#658;.&#618;k/, as follows straightforwardly from (1)? Or must we protect a full (unreduced) short vowel from exposure in syllable-final position, and syllabify as /n&#594;s.&#712;t&#230;ld&#658;.&#618;k/? Perhaps the truth is that speakers differ, and are also inconsistent; but Davidsen-Nielsen&#x2019;s investigation (1974) tends to show that the first type of syllabification, /n&#594;.&#712;st&#230;ld&#658;.&#618;k/, predominates where there is no morpheme boundary after the /s/.</p> <p>The example <i>cacophony</i> /k&#230;.&#712;k&#594;f.&#601;n.&#618;/ confirms that syllable-final short vowels are not absolutely precluded in unstressed syllables; <i>tattoo</i> /&#716;t&#230;.&#712;tu&#720;/, not recorded in EPD with double stress but readily observed with this pronunciation, shows that this licence even extends to syllables having secondary stress. </p> <h3>The affricate condition</h3> <p>The last condition which we have to impose on (1) is one relating to the post-alveolar and palato-alveolar affricates:</p> <blockquote> (5) Affricates (i.e. /tr, dr, t&#643;, d&#658;/) are not split between syllables, but are treated as indivisible. </blockquote> <p>This can hardly be contested in the case of the palato-alveolars. For <i>catching, teacher, allergic, courageous</i> the predicted syllabification again parallels the morphological: /&#712;k&#230;t&#643;.&#618;&#331;, &#712;ti&#720;t&#643;.&#601;, &#601;.&#712;l&#604;&#720;d&#658;.&#618;k, k&#601;.&#712;re&#618;d&#658;.&#601;s/. <i>Ratchet, feature, merger, magic</i> are equally straightforward. Adjusting for place of articulation, <i>petrol, mattress, squadron, Audrey</i> are allophonically parallel; yet the putative /&#712;petr.&#601;l, &#712;m&#230;tr.&#601;s, &#712;skw&#594;dr.&#601;n, &#712;&#596;&#720;dr.&#618;/ might seem to violate the phonotactic condition. It seems, though, that we must accept these syllabifications. If <i>petrol</i> is not /&#712;petr.&#601;l/, what can it be? If it were /&#712;pet.r&#601;l/ we should expect possible glottalling (glottal replacement), as in <i>rat-race, out#right</i>. If it were /&#712;pe.tr&#601;l/ we should have a unique violation in the phonotactic constraint against stressed short vowels in syllable-final position. Given that we demand explicit syllabic boundaries, as we must if the allophonic rules based on them are to be coherent, only /&#712;petr.&#601;l/ remains. This analysis, actually, is supported by the occurrence at the surface level of word-final /tr/ after elision in items such as <i>matter-of-fact</i> /&#716;m&#230;tr.&#601;.&#712;f&#230;kt/.</p> <p>The other phonotactic constraints still apply: <i>district</i> must be /&#712;d&#618;s.tr&#618;kt/, with condition (4) ruling out a possible */&#712;d&#618;st.r&#618;kt/ (which would be wrong both allophonically and etymologically). The rules lead us to choose /&#712;eks.tr&#601;/ as the correct syllabification of the much-discussed <i>extra</i>. The morpheme boundary condition ensures that <i>light-ship</i> is /&#712;la&#618;t.&#643;&#618;p/, and indeed its longish [&#643;] is clearly a token of the fricative /&#643;/, not part of the affricate /t&#643;/. Similarly, <i>board-rubber</i> is /&#712;b&#596;&#720;d.&#716;r&#652;b.&#601;/, although <i>bedroom</i> tends to be pronounced as if morphologically solid, /&#712;bedr.&#650;m/.</p> <p>If it is accepted that <i>petrol</i> is /&#712;petr.&#601;l/ and <i>squadron</i> /&#712;skw&#594;dr.&#601;n/, we must also allow that <i>entry</i> is /&#712;entr.&#618;/ and <i>sundry</i> /&#712;s&#652;ndr.&#618;/. Although this may seem unlikely at first sight, notice the pre-fortis clipping exerted by the post-alveolar affricate of <i>entry</i> on the initial vowel and nasal. And notice the parallelism with the morphology in <i>wintry</i> /&#712;w&#618;ntr.&#618;/. Compare <i>in-tray</i> /&#712;&#618;n.tre&#618;/, with a morpheme boundary separating the nasal from the affricate and no pre-fortis clipping. I would claim that the schwa-elided variant of <i>entering</i> is /&#712;entr.&#618;&#331;/, again with correspondence between syllabification and morphology. Affricates preceded by /l/ must follow the same pattern: <i>paltry</i> /&#712;p&#596;&#720;ltr.&#618;/, <i>cauldron</i> /&#712;k&#596;&#720;ldr.&#601;n/, but <i>mail-drop</i> /&#712;me&#618;l.dr&#594;p/. </p> <h3>Conclusion</h3> <p>I claim that by principle (1), together with codicil (2) and conditions (3), (4) and (5), we achieve a correct syllabification virtually throughout the English vocabulary &mdash; correct, that is, for purposes of predicting appropriate allophones where allophonic variation is sensitive to a syllable boundary. In support of this claim I can report that in the course of working on a new pronouncing dictionary I have transcribed over 50,000 entries with explicit syllable boundaries throughout: indeed, that was the task which led me to formulate the principle. Once I had discovered the principle, it constituted a convenient decision procedure for uncertain cases without, so far as I am aware, any serious untoward results.</p> <h4>Self-criticism</h4> <p>Occasional difficulties do remain. I am worried about words such as <i>accelerate</i> and <i>memorise</i>, where introspection leads me to posit /&#601;k.&#712;sel.&#601;r.e&#618;t, &#712;mem.&#601;r.a&#618;z/ rather than the predicted /&#x2011;&#601;.re&#618;t, &#x2011;&#601;.ra&#618;z/ (compare <i>annotate, advertise</i>, for evidence that these endings normally attract a preceding consonant). Perhaps instrumental evidence will throw light on the accuracy of my intuitions regarding this apparently wayward behaviour of /r/.</p> <p>The distinction between unstressed full and weak vowel is not always clear, since RP /&#618;/ is ambiguous as between these two categories, and so sometimes are /&#601;&#650;/ and /a&#618;/. Are the final vowels in <i>armistice</i> and <i>cannabis</i> weak (/&#712;&#593;&#720;m.&#618;st.&#618;s, &#712;k&#230;n.&#601;b.&#618;s/) or full and therefore consonant-capturing (/&#712;&#593;&#720;m.&#618;.st&#618;s, &#712;k&#230;n.&#601;.b&#618;s/)? RP really offers no grounds for a decision either way. For what it is worth, Australian English does, since Australians would keep a full /&#618;/ as /&#618;/ but turn a weak one into /&#601;/. It turns out that Australians use /&#601;/ in the final syllables of <i>armistice</i> and <i>cannabis</i>, thus indirectly demonstrating RP /&#712;&#593;&#720;m.&#618;st.&#618;s, &#712;k&#230;n.&#601;b.&#618;s/. In <i>politics</i>, as we have seen, Americans treat the final syllable as full-vowelled; so do the Australians (/&#x2011;t&#618;ks/). RP offers no direct evidence, but may be presumed to agree with them.</p> <h4>Ambisyllabicity?</h4> <p>So strong is the presumption among linguists for CV.CV structure as universally preferred that many writers assume it to be true for English even in the face of strong counter-evidence such as is discussed here. Fudge (1984: 21) asserts, on no substantial evidence that I can detect, that <i>competitive</i>, for example, has a stressed syllable /pe/; but American writers, alert to the implications for /t/ allophones, correctly insist on /.&#712;pet./ (see, for example, <i>Webster&#x2019;s Ninth New Collegiate Dictionary</i>). Grunwell (1982) assumes, equally without justification, that a word such as <i>better</i> is /&#712;be.t&#601;/.</p> <p>A more sophisticated idea is that a &#x2018;left-captured&#x2019; consonant such as the /t/ in <i>better</i> is ambisyllabic, belonging to both syllables simultaneously (Kahn 1976: 33; Gussenhoven 1986). This notion has a respectable origin in the phonetic approach to the syllable in terms of sonority: the intervocalic consonant represents a trough of sonority and &#x2018;belongs&#x2019; to neither peak. In modern terms, ambisyllabicity may be felt to allow us to satisfy at the same time both the putative universal preference for CV.CV and the overwhelming allophonic arguments in favour of CVC.V. The principle of Occam&#x2019;s razor, though, shows that ambisyllabicity is not a useful concept. Those who believe in an absolute universal preference for unchecked (open) syllables must, I believe, accept that in English this can at best be true only of deeply abstract representations, and that by the level at which allophonic conditioning becomes relevant a resyllabification rule must have come into operation, namely the principle I propose. And this is uneconomical, since a word such as <i>additive</i>, morphologically /&#230;d+&#618;t+&#618;v/, would have to have been switched to phonological /&#712;&#230;.d&#618;.t&#618;v/ before surfacing again as [&#712;&#230;d.&#618;t.&#618;v]. There may be occasions when the Duke of York gambit is necessary (Pullum 1976), but I do not believe this is one of them.</p> <h4>Acknowledgements</h4> <p>My debt not only to Jones and Gimson but also to Fudge and Kahn must be self-evident, even though I frequently do not agree with them. Gussenhoven (1986) was published only after I had already delivered the UCL staff seminar paper of which this article is a version; I am delighted to see that our thoughts are along the same lines.</p> <h2>References</h2> <ul> <li>Davidsen-Nielsen, N. (1974) Syllabification in English words with medial <i>sp, st, sk</i>. <i>Journal of Phonetics, 2,</i> 15-45 <li>Fudge, E.C. (1984) <i>English Word-Stress</i>. George Allen &amp; Unwin, London <li>Gimson, A.C. (1977) <i>Everyman&#x2019;s English Pronouncing Dictionary</i>. (14th edition.) (Originally compiled by Daniel Jones.) Dent, London <li>Gimson, A.C. (1980) <i>An Introduction to the Pronunciation of English</i>. (3rd edn) Edward Arnold, London <li>Grunwell, P. (1982) <i>Clinical Phonology</i>. Croom Helm, London <li>Gussenhoven, C. (1986) English plosive allophones and ambisyllabicity, <i>Gramma, 10,</i> 2, 119-41 <li>Jones, D. (1931) The &#x2018;word&#x2019; as a phonetic entity. <i>Le Ma&icirc;tre Phon&eacute;tique, 36,</i> 60-5 <li>Jones, D. (1956) The hyphen as a phonetic sign. <i>Zeitschrift f&uuml;r Phonetik, 9,</i> 2 <li>Jones, D. (1963) <i>English Pronouncing Dictionary</i> (12th edn) Dent, London <li>Jones, D. (1977) <i>English Pronouncing Dictionary</i>. (14th edn extensively revised and edited by A.C.Gimson) Dent, London <li>Kahn, D. (1976) <i>Syllable-based generalizations in English phonology</i>. Circulated by Indiana University Linguistics Club. Also published (1980) by Garland, New York <li>Pullum, G. (1976) The Duke of York Gambit. <i>Journal of Linguistics, 12,</i> 83-102 <li>Sharp, A.E. (1960) The analysis of stress and juncture in English. <i>Transactions of the Philological Society</i> <li><i>Webster&#x2019;s Ninth New Collegiate Dictionary</i> (1984) Merriam-Webster, Springfield, Mass.</ul> <hr> <p><div class="small">Placed on the web by <a href="mailto:j.wells@phon.ucl.ac.uk">J.C.Wells</a> 2002 12 11</div></p> <p><div class="small">my <a href="http://www.phon.ucl.ac.uk/home/wells/">home page</a></div></p> </body> </html>