Is the phoneme dead?
We’ve been doing a unit on phonology, and my doubts about the phoneme are partly a reflection of my students’ own difficulties with the concept. Not surprisingly, I’ve been having to tease out the difference between phonemic symbols and phonetic symbols, and even between phonology and phonics.
But all the time I’ve been dreading the day when someone challenges this definition (from An A to Z):
‘A phoneme is one of the distinctive sounds of a particular language. That is to say, it is not any sound, but it is a sound that, to speakers of the language, cannot be replaced with another sound without causing a change in meaning’.
The definition has an authoritative ring to it, not least because it simply re-states what by many is considered a founding principle of all linguistics. Listen to Jakobson (1990: 230) who practically bellows the fact: ’The linguistic value … of any phoneme in any language whatever, is only its power to distinguish the word containing this phoneme from any words which, similar in all other respects, contain some other phoneme’ (emphasis in original).
How is it, then, that we regularly teach that the ‘s’ at the end of cats is a different phoneme than the ‘s’ at the end of dogs? If different phonemes flag different meanings, what change of meaning is represented in the difference between /s/ and /z/? Or, for that matter, between final /t/ and final /d/, as in chased and killed? If there is no difference in meaning (since /s/ and /z/ both index plurality, and /t/ and /d/ both index past tense), aren’t they simply different ways of pronouncing the same phoneme?
Phonemes, after all, are not phones, i.e. sounds. Acoustically speaking there are many different ways – even for a single speaker – of realizing a specific phoneme. This is why Daniel Jones (1950: 7) defined phonemes as ‘small families of sounds, each family consisting of an important sound of the language together with other related sounds’ (my emphasis). These related sounds are the different allophones of the phoneme.
Hence the analogy with chess pieces: the way individual chess pieces are designed will vary from set to set, but they will always bear certain family resemblances, bishops all having mitres, and knights having horse heads, etc. More important than their form (and one reason that this analogy seems to work so well), is the relationship that they have with one another, including the ‘rules’ that constrain the way that they may behave. Bishops can’t do what knights do, nor go where knights go, and vice versa.
Phonemes – like chess pieces – are defined in relation to one another. As Bloomfield (1935: 81) put it, ‘the phoneme is kept distinct from all other phonemes of its language. Thus, we speak the vowel of a word like pen in a great many ways, but not in any way that belongs to the vowel of pin, and not in any way that belongs to the vowel of pan: the three types are kept rigidly apart.’
In fact, a purely structuralist argument would say it’s not actually about meaning at all, it’s about ‘complementary distribution’, or, as Jones (1950: 132) puts it (also bellowing): ‘NO ONE MEMBER EVER OCCURS IN A WORD IN THE SAME PHONETIC CONTEXT AS ANY OTHER MEMBER’. That is to say, the /s/ at the end of cats and the /z/ at the end of dogs never occur where the other occurs, and vice versa. But is this true? What happens to the /z/ at the end of dogs in the sentence: The dogs seem restless? Hasn’t it become /s/?
Ah, yes, you say – but sounds in connected speech are influenced by their environment, blending with or accommodating to the sounds around them. The true test for a phoneme is if it distinguishes isolated words, like pin and pen – those infamous minimal pairs. But when are words ever isolated? When does the phonetic environment not have an effect? And isn’t the voiced /z/ at the end of dogs, and the unvoiced /s/ at the end of cats also an effect of the phonetic environment? That is to say, where does connected speech start becoming connected if not at the juxtaposition of two sounds?
It gets even trickier when we consider weak forms. There are at least two different ways of saying can, as in I can dance: I /kæn/ dance, or I /kən/ dance. Both are possible, even where the stress remains on dance. The latter is simply more reduced. But the meaning is unchanged. [kæn] and [kən] are not minimal pairs. They are different phonetic realizations of the same word (hence the square brackets). Phonetic. Not phonemic. Shouldn’t, therefore, they both be transcribed as /kæn/?
In researching this, I’ve encountered a lot of debate as to whether the concept of the phoneme has any currency at all any more. As one scholar puts it, ‘the phoneme, to all appearances, no longer holds a central place in phonological theory’ (Dresher 2011: 241). The problem seems to boil down to one of identification: is the phoneme a physical thing that can be objectively described, or is it psychological – a mental representation independent of the nature of the acoustic signal?
The answer to the first question (is it physical?) seems to be no, there are no ‘distinctive features’ or family resemblances (such as voicing or lip-rounding) that unequivocally categorize sounds as belonging to one phoneme family and not another.
On the other hand, there is some evidence, including neurological, that the phoneme does have a psychological reality, and that speakers of languages that share the same sounds will perceive these sounds differently, according to whether they flag meaning differences or not. (This is analogous to the idea that if your language does not distinguish between blue and green, you will see both blue and green as being shades of the same colour). This, in turn, is consistent with Jakobson’s claim that ‘if we compare any two particular languages, we will see that from an acoustic and motor point of view their sounds could be identical, while the way they are grouped into phonemes is different’ (p. 223).
It’s not for nothing, therefore, that the concept of the phoneme has given us the very valuable distinction between emic and etic, i.e. the perspective of the insider vs that of the outsider. Phonemes capture something that we, the insiders, intuit about language, even if their objective reality is elusive. We know that pronunciation impacts on meaning, even if we don’t quite know how.
Perhaps Jakobson (op. cit. 230) had good reason to claim, therefore, that ‘the phoneme functions, ergo it exists’.
References:
Bloomfield, L. (1935) Language, London: George Allen & Unwin.
Dresher, E. (2011) ‘The Phoneme’, in van Oostendorp, M., Ewen, C.J., Hume, E., & Rice, K. (eds) The Blackwell Companion to Phonology, Oxford: Blackwell, available here
Jakobson, R. (1990) On Language, edited by Waugh, L.R. & Monville-Burston, M., Cambridge, Mass: Harvard University Press.
Jones, D. (1950) The Phoneme: Its nature and use, Cambridge: W. Heffer & Sons.
Illustrations from the very clever phonemic chart that comes with English File (Oxenden, C. and Seligson, P., 1996, Oxford University Press).
An A-Z of ELT 
Hardcore pron is a bit more than I can handle this early on a Sunday morning, but thanks for the timely reminder that I should revisit English File, a course book that was very much ahead of its time in one or two very worthwhile ways.
Thanks as ever for an illuminating post, Scott*. I’m looking forward to reading others’ thoughts on phonemes.
Meanwhile, I think I might write a retrospective review of Oxenden’s and Seligson’s series. I’m sure you didn’t imagine taking that as their inspiration from this post!
*Sometimes it should be enough to say thanks, I feel.
Thanks Adam; re English File, it was a bit naughty to post those bits of their chart, but I hoped they would consider it good publicity (as if they needed it!), and I deliberately didn’t post the whole chart. Your positive review would help mitigate my trespass, hopefully!
Where do allophones fit into the picture? Is it right to say they are different pronunciatons of a central phoneme but still classified as acceptable. And if that is so would one have to define a phoneme as a group of sounds, all of which convey the same meaning…. or someting like that. ??? Dennis
Yes, Dennis, allophones are the different variants within a phoneme ‘family’. To give the Jones (1950: 7) quote it’s full due:
So, the clear [l] in like is pronounced differently from the dark [ɫ] in milk, but if you were to use a clear [l] the milk would still be milk. That is to say, the allophones don’t impact on meaning.
Hi Scott,
Enjoyable post. I always understood the phoneme to an ideal abstraction of a possible sound and the smallest phonological unit that could change meaning, hence minimal pairs.
I remember doing phonology at uni and learning about different realisations of phonemes being called allophones. The classic example I remember from uni was the distinction between dark and light “L”, leading to allophonic variation in English with /l/ being realised as [ɫ] “dull” or [l] “lip” in a relationship of complimentary distribution. This also highlighted the language-specific nature of these things, as the two allophones in English could be separate phonemes in other languages. The same would apply for aspiration, such as with /p/ being realised as [p] or [pʰ] depending on phonological environment. Is that right?
The differences you highlight between “cats” /s/ and “dogs” /z/, along with “buses” /ɪz/, were what I learned as allomorphs (I think – we are going back a bit here) and so the distinction would be /s/ – [s], [z], and [ɪz]. The same would apply for “-ed”. In fact, if I remember rightly, this also led to the (remarkable) analysis that “sheep” singular turned into “sheep” plural with a change of [0]. Do I remember that correctly?
Taking into account all of that, does that mean that /s/ and /z/ would be analysed as distinct phonemes in, say, “zed” and “said” (in my accent at least) but as allomorphs of the abstract plural marking suffix /s/ in the examples above?
Anyway, thanks for the though-provoking post!
Chris
Oops, thanks Chris, your comment answered Dennis’s question about allophones better than I could (it seems that the light /l/, dark /l/ distinction is still the most popular exemplar!)
Yes, my sources give ‘allomorph’ as a variant of a morpheme – so a feature of word formation, rather than of phonology per se, although a division is made between those that are realized in the pronunciation (as in the three plural allomorphs [s], [z], and [ɪz]) and those that are realized in a change of both pronunciation and spelling (as in swim/swam). I would call your sheep/sheep a homomorph, by analogy with homonym. But I just invented that!
Hi Scott,
Thanks for an interesting post.
Without wishing to be too cheeky in my first ever reply, I think that the definitions that you and Jakobson give look at the phoneme as the sole method of distinguishing two otherwise identical words. Given that authentic language use is always situated and contextual, this would leave phonemes as just another tool for meaning making, rather than the be all and end all. If you follow the definitions to their logical end points, you end up, as you did in the post, at a minimal pairs exercise, where phonemes are the only distinguishing feature, rather than one of a group including lexical, grammatical and discoursal clues. In my teaching experience, pronunciation errors cause misunderstandings mostly when my students are trying to give me decontextualized words.
I’d say the key phrase in Jakobson’s definition is “power to distinguish”, which I might recast as “potential to distinguish”. That would mean that phonemes don’t necessarily change meaning every time, as in the case of past tense endings. It might also leave a bit more space for non-standard pronunciations such as in Jenkin’s ELF core, and her example of /th/ being substituted for /d/ and /t/. In this case, changing a phoneme quite clearly doesn’t change the meaning in ELF contexts (according to Jenkins). I wonder what this does to Jakobson’s point about functioning?
Cheers,
Alex
Thanks Alex – well put! Your comment stated better than I could why the notion of the phoneme might need to be quietly interred. The fact that meaning is contingent on a much wider – and less discrete – range of factors suggests that an obsession with phonemes may be counterproductive. Nevertheless, it has led to some quite extraordinary claims, such as Anthony Burgess (in Language Made Plain):
Excellent as usual, Scott; thank you. However, I wondered why in your referencing of the structuralist argument you don’t give a nod to Ferdinand de Saussure? I’m less familiar with Bloomfield, but his chess analogy above is straight out of Course in General Linguistics* (well, in essence, anyway). The very idea of a grid of meaning based on the relational value between units of a group–i.e. that “pen” means pen because it doesn’t mean “pin” or “fen” or “pet”–originated with Saussure, if I’m not mistaken. Even your closing points regarding “mental representations independent of the nature of the acoustic signal” seem to invoke the ghost of Saussure’s langue vs. parole.
Cheers,
Marcos
* A truly remarkable book in several ways–and a gripping read too, despite the unlikely title!
Thanks for the post, Scott.
Based on the conversations I’ve had with other teachers, it seems that studying phonetics, never mind phonemics, gives trainee ELTs some of the biggest headaches!
Just on a simplified classroom teaching level, personally I’ve always been fascinated by how a seemingly minor (and, of course, unexpected) variation of sounds in a word or phrase creates utter confusion for the listener. When (after the confusion) I give the student the more ‘native-like’ pronunciation, I often feel quite guilty that I didn’t catch what they said the first (or second or third) time, when in retrospect the pronunciation difference usually seemed to be graspable. Naturally, there are a number of issues at play with phoneme/stress/intonation etc, but none-the-less it seems incredible how finely-tuned much pronunciation is (or at least, how finely-tuned our expectations are).
Thanks again 🙂
Thanks Avi. Check out Richard Caudwell’s approach to helpig learners deconstruct the stream of speech: http://www.speechinaction.org/
(I notice, by the way, he’s got a book coming out imminently – should be worth getting hold of).
Hi Avi,
I appreciate I’m joining this conversation a few years late! I am a CELTA/Trinity Cert TESOL trainer in the UK and I’ve developed an online moodle course to support my trainees.
If you have a spare moment, I wonder if you wouldn’t mind taking a look at the latest trial unit available on the site called ‘Sounds of English’. It’s a phonology unit designed to help learners understand the phonemic chart and recognise the symbols and their related sounds. It includes interactive phonemic tables with lots of interactive tasks to help check and consolidate learning. There is no need to log in, simply click the following link to view the unit: http://moodle.train-to-tefl.com/course/view.php?id=13
I would love to know what think.
Suzannah, I’ve had a quick look (sorry, I don’t have time to go through it all!), but I would say it all looks pretty logical and comprehensive. I don’t know, but maybe the presentation and display format can be jazzed up a bit, and you might want to check about ease of use with your students? Otherwise, it seems pretty sensible as a general resource.
Nowadays, to be honest, I tend to use my own personal style of modeling and exercises with students. I don’t often refer as much to such resources, I guess, as I’ve built up my own methods which seem to ge the job done.
Thanks, Marcos. Yes, my mention of the contrastive nature of the phoneme being ‘one of the founding principles of linguistics’ was a nod to Saussure, specifically to the idea that ‘every sign in a system derives its values from its difference from the other signs in that system’ (from K. Malmkjaer, ed. The Linguistics Encyclopedia, Routledge, 1991, p. 438).
From this basic premise grew the Prague School approach to phonology, (further developed by Ramon Jakobson) which (and now I turn to David Crystal) ‘defined the phoneme as a bundle of abstract distinctive features, or oppositions between sounds’, whereas Daniel Jones ‘viewed the phoneme as a “family” of related sounds, and not as oppositions’ (A Dictionary of Linguistics and Phonetics [5th edn], Blackwell 2003).
Who would have thought the poor little phoneme should bear such a burden of theory!
As an L2 learner myself, my perspective is that the phoneme is not dead as long as we understand sound-production ability as belonging to a more socio-cognitive (or cognitive-social?) construct rather than purely structuralist approaches to phonology and phonetics. In the long run, concepts such as the Ideal L2 Self, language ownership, private speech, fluency as automatization, or ZPD may be more productive in instructed language learning settings.
A lot of new information to chew on, in both the post and the comments. Thanks. Like Avi, I’m looking at it from the classroom level.
I’m leaning toward agreeing that the concept of phonemes may not be so useful. My sister-in-law, who is from Kentucky, says something like “pin” when she’s talking about a pen. It can be a source of miscommunication if there is not enough context, but usually that’s not the case. And I think my brain tunes itself to consistently-used deviations from what I’m used to hearing. Once I’m tuned, “pin” and “pen” can have the same meaning (depending on who’s speaking and what we’re talking about). I guess that’s evidence of a mental representation?
Yes, Kathy, I think ‘tuning’ is a good metaphor – and maybe relates to the notion of accommodation (dealt with in an earlier post), whereby we adjust (re-tune) our mental schema to deal with the new input. Which makes me wonder if there are prototypical sounds in every language, familiarity with which helps align the whole system. (Like the way the whole orchestra tunes to just one or two notes at the beginning of a concert, as modelled by the lead violin).
Yes, i was wondering something similar. Maybe when there are two languages involved it’s sort of like a musician who is well-informed in her own musical system trying to play with musicians who use a different system. She may not have to switch over completely, but rather understand how the systems differ and maybe find resonances between them. Just for fun (maybe there’s a parallel?):
From a pedagogical point of view, I hope that the phoneme is not dead. I’m a firm believer that a well developed ability to recognize patterns (and changes in patterns), be they sounds or symbols, is a indispensable language learning skill. I don’t find time spent on them to be counterproductive at all and neither do my students. Though I do have to admit that the phoneme is a very abstract concept, and that learning about them always seems to give both teachers and students trouble, I still love the idea that there is some kind of attainable standard (not as precise as tuning an instrument to 440 Hz, but one that promotes comprehensibility). My point of contention with introducing such an abstract concept is the set of symbols we use to represent them. While those who teach in areas where the market dictates the socio-cognitive construct of (RP) version of British English accent as the model benefit from a more or less united sound-symbol representation system, come on over to the U.S. and you’ll see how all of our favorite publishers, Cambridge, Pearson Longman, Macmillan, and Oxford among others, have done little to promote the teaching of pronunciation as an integrated skills in textbooks or to adopt a standard for representing the sounds. While the heads of departments in community colleges in my area sit around arguing over which symbol chart to use for teaching, I’ll take English File chart any day!
And by the way, the community colleges in my area chose the one for RP, as used by the British Council! Why do you reckon that happened?
Thanks, Delpha – I share your dismay at the lack of a standardized and pedagogically oriented list of phonemic symbols in the US – not least because I have to teach phonology there in the summer! I’ve tried to come up with my own solution – you can read about it in this post:
P is for Phonemic chart
Yes, I remember that post. Thanks! It was brilliant and I have printed and used your chart in my classes on several occasions. Now, if we could just get the nation and the publishers on board!
Scott, just emailed you a UK/US phoneme chart my CELTA tutor gave me way back when. British Pound symbol next to /ei/ and U.S. dollar sign next to /e/, for example. How arbitrary is that? It might be fun to add it to your collection of exemplars, the way my DELTA tutor once collected authentic lesson plans by gathering scraps of paper and odd notes from teachers working at the IH Center that day.
I like to think of phonemes much like Plato’s Forms (Ideals): the archetype of a horse, for instance, provides the Form by which we can recognize, identify, and imagine a horse. An animal like a donkey then becomes horse-like but different, warranting it’s own name (in US English anyway). Without these forms and the archetypes we inherit as humans, we would have no imaginal world to draw upon.
Along with RP and other phonological wonders comes the notion that there is a correct way to sound, which leaves my Spanish speaking students at a disadvantage every time they hear the ‘s’ at the end of ‘cats’ but I don’t, or when they place /s/ at the end of most every plural noun. As has been mentioned, context is key to helping us understand ‘green bottles’ even when we say ‘greem bottles’ (try it three times fast), so why not make room for other Englishes besides ‘air hair lair there’ (My CELTA tutor’s introduction to RP: say it without the final ‘r’ if you wouldn’t otherwise, to produce ‘Oh hello there’ :-).
I was one of those nerds who understood and used phonemes from the get go on my CELTA while otherwise successful candidates struggled with ‘another alphabet’, as some language learners have called the phonemic chart. I was once hired to train pre-service teachers on a four-week course in Greece that placed phonology at the core of the curriculum. I’ve heard it said phonemic script should follow every boarded lexical item, that it encourages learner autonomy by making more sense out of learner dictionary entries. The phoneme doesn’t seem dead just yet.
And I introduce learners to phonemes sometimes, but always with the understanding that their English(es) will sound different than mine but be effective enough to communicate with and comprehend other Englishes, as long as the phonemes work within context. If it looks and quacks like a duck, it’s probably a duck?
Rob (who says ‘pin’ when he means ‘pen’)
You’re a pin-sayer too? I will tune my reading of your posts accordingly!
😉
One of my students ‘corrected’ my pronunciation of pen this term.
Thanks, Rob! Regarding archetypes, the Dresher (2011) chapter I refer to does mention the view (associated with cognitive linguistics) ‘in which phonemic representations are prototypes or basic level categories abstracted from lexical representations’ (p. 250). Which reminds me that connectionist models of neural networks, ‘in the course of processing particular exemplars, … often acquire knowledge of the underlying structural regularities in the whole problem space. They develop representations of categories and prototypes. They generalize from this knowledge’ (N. Ellis, 2003 ‘Constructions, chunking & connectionism’, p. 87).
Connectionist models suggest that the basic unit of phonology is not the phoneme, but the syllable. In our L1 we acquire a ‘self organizing map’ of the syllable system (MacWhinney 2008), which is rehearsed and entrenched through babbling. This syllable map sets the range and limits on what sounds can go with what. Japanese, for example, has a relatively small syllabary consisting of around 70 items. English, on the other hand, has as many as 3000 possible syllables. The Japanese problem with /l/ and /r/ may be more to do with not having syllabic environments for these sounds, rather than the sounds themselves.
Needless to say, restructuring a syllable map for a second language requires enormous effort. ‘L2 learners must rely on repeated focused trials to link changes in the auditory syllabary [i.e.the syllables they hear] with changes in the articulatory syllabary. Methods for inducing these changes include presenting clear cases, facial visual feedback, and diagrams of tongue positions’ (MacWhinney; B. 2008, ‘A unified model’, in Robinson & Ellis, A Handbook of Cognitive Linguistics in SLA). No mention of the phonemic chart, by the way!
PS: Plato’s Forms, or Ideas (not Ideals) -sometimes too dark to see in this cave. 🙂
That chart from E. File was handed out to lucky delegates at Glasgow and it has helped me to learn some of the IPA symbols that I had never managed to remember before. It was the most useful item I took home with me (Now, there’s a thought!). Now when I see the symbol “e”, I can visualise that “egg” and my students know the pronunciation of, for example, “any” or “friend”.
I’m afraid my understanding of and employment in teaching of phonetics doesn’t go any further than that.
It might well be off-topic, but is there a reason to do with phonetics that Peking is now called Bejing?
Thanks, Di – I envy you your free English File chart! A clarification, though: the symbols on the chart are only a small sub-set of the IPA symbols, i.e. those that in English are phonemic, in that they are the smallest units that account for differences in meaning. Hence it’s a phonemic chart, not a phonetic one. Whereas a phonemic chart of RP has 44 symbols, the IPA chart has a least 100 plus a pile of diacritics that can be attached to them.
As for Peking/Beijing, the change results from the adoption of the pinyin system of transcribing Chinese into roman script, a system considered more accurate than those that preceded it. It is phonetically-grounded, but inevitably approximate, and does not map on to the typical English pronunciations of the different letters in every case.
For a really interesting chart of correspondences between pinyin and the IPA, see the relevant entry in Wikipedia: http://en.wikipedia.org/wiki/Pinyin
Thank you, Scott. Now I know where to look up the Peking/Beijing thing!
Jakobson (1990: 230) who practically bellows the fact: ’The linguistic value … of any phoneme in any language whatever, is only its power to distinguish the word containing this phoneme from any words which, similar in all other respects, contain some other phoneme’ (emphasis in original).
I think a much better example of what Jakobson is trying to say is the words close and close as in
“He´s a close friend.” and “Please close the door.”
but there are many other examples of this minute, excuse, wind, etc
“cats” and “dogs” are not simillar in all other aspects.
Thanks, Alina – yes, I didn’t mean to suggest that cats and dogs are minimal pairs – clearly they’re not (cf caps/cats and togs/dogs, etc). I wanted to suggest that the choice of /s/ or /z/ is a phonetic one, not a phonemic one, since meaning is not affected: the addition of /s/ or /z/ makes the noun plural in each case. (It gets trickier when you have a noun like horse, where pluralization adds, not just a sound, but a syllable.
By the way, the pairs that you cite (close/close, etc) are all homographs (i.e. words spelt the same but pronounced differently) but it’s not the case that minimal pairs have to be homographs, is it? Viz caps/cats.
Truly interesting post Scott, classes have just begun in Argentina, where I’m teaching right now, and I’ve been experimenting with the phonemic chart, teaching it to all my learners, regardless of their age / level (from kids to adults, int to adv). So far, it’s been an empowering tool, and of course I don’t focus on tiny things like the s or z to indicate plural, I focus more on meaning, and how an awareness of these sounds can help them indirectly improve their speaking and listening. I think Underhill’s tips regarding what goes on in your mouth when you produce these sounds, has been really useful, as u’ve said before, lip rounding towards u. u: o:, etc, smile (ish) for i:, etc, these tiny things have helped my learners to start thinking of how they produce sounds, and to at least consider that it’s not the same vowel sounds we have in Spanish, some are more similar than others, but .. so I would favour phonemes as an abstract representation, if we use it for good methodological pruposes and to get our learners aware of a different sound system… when they learn the schwa, and when they understand stress time and play with stress patterns as well as different sounds and when they focus on their mouth and how they are producing a sound, I feel happy. And by challenging myself to try out new things, I’ve discovered it’s not impossible, and learners don’t find it that hard, the phonemic chart is teachable and I’d say useful to a certain extent. As it’s been stated above, some books get learners to experiment when it comes to hypothesis about meaning or form of language, but rarely do we get people to experiment with sounds, so I guess, it’s up to, teachers, and I wouldn’t focus on the s or z for plurals, because it doesn’t change meaning, but I would focus on things like the letter Z in Ame (zi:) and the sea (si:).. those minimal pairs that change meaning or if we are learning a new word, or expression, discuss how we pronounce it, not just repetition, but awareness…
Thanks for sharing, and Delpha, I agree 100% with what you’re saying.
.…so I would favour phonemes as an abstract representation, if we use it for good methodological purposes…
Yes, indeed, Nati – I tend to agree. Thanks for your comment and apologies for the delay in replying. I particularly like the way Adrian stresses the sheer physicality of pronunciation, something that he himself commented on here.
Great post on B is for body. thanks Scott!
Macmillan education posted Underhill’s workshop on successful pronunciation…. and this has to do with what he says here and on the post you’ve just showed me. http://www.youtube.com/watch?v=1kAPHyHd7Lo&playnext=1&list=PL1EBF24E4DE6085C5&feature=results_main
Scott,
Phoneticians are wholely concerned with the pronunciation of words. Users of a language are nearly as concerned only with understanding which words they are saying or hearing and, consequently largely ignore such phonetic niceities in their own and others speech. I propose,then, the following definition of phoneme for them:
Phoneme (as speech element) — Phonemes are the fundamental units of a language that, spoken or written in sequence, form the words of that language.
Refer to “Phonemes, their significance and use” in http://www.akses.org/phonemedef.htm
Jim Kanzelmeyer
Thanks, Jim – yes I agree, that even if the concept of the phoneme is problematic, and the reality elusive, there is still good sense – from a teaching point of view – of restricting the phonetic ‘burden’ to those sounds that, intuitively to a native speaker, impact on meaning, and hence (in all likelihood) on intelligibility. (Although, even here, there are fewer phonemes than the full 44 that are probably worth approximating, if the evidence from ‘phonological core’ studies can be taken seriously).
A fascinating area – so much so that I do all the pron sessions on our CELTA courses.
Some thoughts (and apologies if I simply re-state something said earlier):
1. The schwa as our default vowel, and allophone of most of the monophthongs, seems to me to make this system of pure vowels more closely-knit as a family while at the same time being more labile – probably represented as a constellation of relations that is, to a smaller or greater extent, reconfigured on meeting a new accent (I recall this reconfiguring process in the first five minutes of listening to the Sheffield accent in the film The Full Monty). I agree that this relation is perhaps held very abstractly much like a melody is recognizable when played by different instruments, as mentioned above.
2. As for consonants, this is where perhaps Prototype Theory (which you allude to) could be put to work since consonants are more pindownable (?) as a combination of a sets of features (e.g. voiced/devoiced). Here the central phone (as you pointed out) is our ‘sparrow’ [I am thinking of the diagram from Words in the Mind by Jean Aitchison ] while ‘ostriches’ and ‘kiwis’ lie in the outer orbits where they may start to intersect with orbits from ‘neighbouring’ phoneme clusters, proximal on the phonemic chart (based as it is on the vocal tract space).
3. The phoneme box probably contains all the allophones including those formed through sandhi variation.
4. As John Field in Listening in the Language Classroom suggests, when we hear a word like “veshtables” – we probably move into the Meaning realm more consciously, and use our knowledge of primings, topic and discourse structure to salvage the mispronounced phoneme. Then go back to decoding simultaneously on a Meaning-Form(Pron) dual track process. {Not sure if this observation is relevant to the discussion !}
5. Rather than learning 3000 syllables and connecting these, isn’t it also possible that the neurons learn more abstract rules of phonotactics (such as the sonority hierarchy)[http://en.wikipedia.org/wiki/Phonotactics] . In a way, these are sets of ‘primings’, similar to lexical primings (a al Hoey) and it is perhaps these two sets of primings, phonemic and semantic, which could be the ultimate meaning-form relationship represented in the mind.
Thanks for this very interesting comment, Declan. Coincidentally, i was discussing these issues just yesterday with one of my students (who happens to know a lot more about phonetics than I do) and we both seem to have arrived at a similar point, regarding the schwa less as a phoneme in its own right than as an allophone of virtually all the other monophthongs.For a start, it seems odd to have only one phoneme that is characterized as always unstressed, where no other vowel is described in terms of its context like this. For another, the minimal pairs it distinguishes a very few: conquer, concur being one – again, the difference is in stress as much as in any intrinsic vowel quality. So maybe we should remove the schwa from the phonemic chart altogether, or, if not, conflate it with the NURSE vowel, as an allophone.
At the same time, the schwa has enormous importance in terms of its position as indicating the ‘neutral’ position in the English articulatory setting. Arguably, aligning the articulators so that the schwa is central may be an important way of ‘unlocking’ the whole system.
Regarding your 4th point, about phonotactics, I’m totally in agreement, and Nick Ellis has written about the way that from birth we are ‘primed’ to recognize and reproduce the phonotactics of our first language as the first stage in abstracting the permissable sequences of the language.
All this is in haste – I’m at the TESOL convention in Dallas at the moment, and looking forward to a session on articulatory settings tomorrow. I hoep to report back soon!
Scott,
The conversation has moved on, but I wanted a chance to chime in, because I believe at TESOL I’d mentioned a couple of the authors/theorists I think of when this topic comes up.
First, let me say that I think that the concept of the phoneme is a useful one, both as short-hand in certain kinds of phonological analysis, and as a pedagogical tool. The concept of speech sounds is both familiar and mysterious, particularly when we’re trying to recognize, classify and reproduce sounds that may be foreign to our understanding and perception. Any guidepost, particularly one that has some demonstrable theoretical basis, and such extensive usage over the last century, deserves consideration.
Sapir’s The Psychological Reality of Phonemes (1933), Swadesh’s The Phonemic Principle (1934), Bloch’s A Set of Postulates for Phonemic Analysis, and Pike’s Phonemics (1947) — the last was one of my textbooks for introductory phonetics even in the early 90s, still being reproduced in all its original typewritten goodness — are my main starting points for understanding the phoneme, as a purely mental construct, which is manifested in sound. Bloch and Pike do the most in terms of making the phoneme a usable tool, with which we can organize the meaningful sound ranges of a language. What strikes me now, looking back at some of their writing, is how close they seem in some ways to showing the kind of complex associations that we might expect from modern neuro- and psycholinguistic descriptions of phonology. They each lay out patterns of triangulation that allow us to identify a particular segment as belonging to one phoneme and not another.
Of course, the concept has its flaws, and schwa, as well as the medial alveolar flap neutralizing /t/ and /d/ (writer/rider) in NAE pose problems for the orthodox view of the phoneme, since they each would clearly have to be allophones of more than one phoneme by traditional analysis, and thus violate biuniqueness. Trubetzkoy came up with the concept of the archiphoneme to deal with just this issue. An archiphoneme (usually represented by a capital letter, just as we generally use D to represent that pesky flap when teaching it to students), is a sort of underspecified segment that adapts to its surroundings. This follows from the concept that phonemes are primarily psychological concepts, holding places in a phonological pattern. This would definitely apply in the case of /t/-/d/ neutralization. Whether this would clarify anything for learners is doubtful, as it would still leave them wondering if it was /t/ or /d/, ‘writer’ or ‘rider’. Fortunately, context will almost always sort that one out. It might be more difficult to use in explaining schwa, since here we have not just two phonemes competing for the same allophonic realization, but several of them.
I think, perhaps, that it would help to introduce the idea of schwa as “neutral ground”, and not a separate sound fairly early on, as is done when teaching Russian, which has its own similar vowel alternation in a two-part pattern in unstressed syllables. Schwa is simply what many unstressed vowels sound like, and we can point to examples and patterns for learners to examine and practice. What I realized after thinking this over after a discussion in your class, is that if we hear the same sound in a stressed position, we know it is not schwa, but /ʌ/, despite the fact that they sound the same in many dialects of English. We know this, not necessarily strictly on the grounds of one segment versus another, but how the segments in a word function in the words it’s related to. Thus, we know through all the exposure and establishing of neurological links that the second ‘a’ in actual and actuality is realized by schwa in the first word, and not in the second. Although we like to concentrate on the idea of phonemes as discrete units, their reality is so much more complex and interdependent, as they help weave the sound of the language.
It’s great that you managed to respond to my comment in the whirl of a conference. You extended the notion of the schwa as a common allophone to arguing it out of existence in the chart altogether (!) which I agree with from a technical point of view; but from a pedagogic viewpoint (which we also hold) it is perhaps best kept there, if only to create as you say an indication as to
“the (central) ‘neutral’ position in the English articulatory setting, …(which can) ‘unlock’ the whole system”. This is exactly how I use it in input sessions (inspired by Adrian Underhill) for trainees to get an anchor within that grid of vowels.