WARNING! If you can see this paragraph, then you are probably using Firefox browser in its “Reader View”. Firefox says this is to remove clutter, but what it judges to be clutter is the linked audio, which is an essential feature of this essay. Please exit Reader View.
Around 1980 I stayed for a couple of nights in a hotel in Birmingham, England, about 200 miles to the south of my native Scotland. Feeling thirsty, I asked at the public bar for a Coke. Off went the barmaid and she returned promptly. But she was not bearing the Coca‐Cola that I expected. “Is this what you want?” she asked cheerily. And she presented me with a cork.
I was amused and surprised. I articulate my speech clearly. In my speech the words Coke and cork are very different. The vowels are not even close to each other, and, like any Scot, if I say an ‘r’ you’ll hear it. Yet I was misunderstood, and I was misunderstood because Scottish English speech exhibits fundamental differences from all other varieties of English. This essay is a celebration of Scottish English speech and an exploration of these differences.
Setting the scene
First let me explain the geography for the benefit of any reader who does not know the UK. By far the largest constituent of the United Kingdom in terms of population is the nation of England, with 53 million inhabitants. To its north, England shares a common border with the nation of Scotland, which has 5 million inhabitants. To its west, England shares a common border with the nation of Wales, which has 3 million inhabitants. Across the sea from Scotland to the south‐west is Northern Ireland, with 2 million inhabitants. The UK consists of these four parts, and the predominant language in all four is English.
Scottish English refers to any of the varieties of English spoken by the people of Scotland. Within Scottish English there is of course a lot of variation. There is variation by region, by social class, by age, by gender, and in places even between Protestant and Catholic. But the major variation is not primarily by any of these. The major variation is along a “bipolar continuum” – a line with two ends – with Scottish Standard English at one end and focused Broad Scots at the other. Scottish Standard English or SSE is the formal, prestige, variety of Scottish English: it is the characteristic speech of the professional class in Scotland and the accepted norm in schools. It is also by far the easiest variety for outsiders to understand. The other end of the continuum, Broad Scots, tends to be informal, varies more by region than SSE does, and is notoriously impenetrable to outsiders. Many Scottish speakers are able to code‐switch along the continuum to fit the social situation that they are in or that they want to create. This essay is primarily about Scottish English as a whole; where I consider variation within Scottish English, I mainly consider variation along the Broad Scots ↔ SSE continuum.
Is Broad Scots a dialect of English or a distinct language? That question has a linguistic answer and a political answer, and the two answers are different. Linguistically, Scots used to be a separate language, evolved from Northumbrian Old English around 1100 AD, but over the 300 years since the Union of the Parliaments in 1707 it has become so diluted by contact with modern English that most linguists now regard modern Scots as a dialect of English. I will follow the majority linguistic view and call it a dialect. But there is a quip that a language is a dialect with an army and a navy. God forbid that Scotland might ever have an army or a navy, but the essence of the adage is that a dialect will be politically regarded as a language if there is sufficient national will for it to be so regarded, and in this case there is ample such will. Since 2001 the UK government has accepted Scots as a language and has recognised it as such under the European Charter for Regional or Minority Languages.
The Scots dialect has nothing to do with Scottish Gaelic. English, with Scots, is a Germanic language, whereas Scottish Gaelic is a distantly related Celtic language which only about 1% of the population of Scotland can speak. I cannot speak or understand any Scottish Gaelic.
SSE can stand for “Scottish Standard English” or “Standard Scottish English”. I prefer the former name. It emphasises that SSE is not merely a variety of English: it is a variety of Standard English. It sits under the umbrella of Standard English as surely as does the Queen’s English, or General American English, or Australian Standard English. The reason that SSE is called a “variety” of English is not that it varies from anything. In linguistics, a language variety just means a specific form of a language or language cluster. Everyone who speaks English speaks a variety of English.
Scottish English differs from English English (the English language as spoken by the people of England) in several ways. It differs slightly in its lexicon, its stock of words. It differs slightly in its grammar, the way that words are put together. But by far the biggest difference is in its accent, the way that words are pronounced. This essay is about accent. Like variety, the word accent in linguistics does not imply a deviation from any standard: a person’s accent simply means the way he or she pronounces words, and a person who speaks the most Received Pronunciation has an accent as surely as I have. We shall see in this essay that Scottish accents differ from accents of England in ways that are not merely large in size: they are fundamental.
I am from Edinburgh, in the east of Scotland, and I speak Scottish Standard English. I can’t code‐switch. Click on the green text below to hear a sample of my speech, which I trust you have no difficulty in understanding.
Which way should we go to Lochwinnoch? One way is seven miles: the other isn’t quite so far, but I don’t want to take the car on that bad road again.
Here is the same passage in another middle‐class SSE voice, this time from Renfrewshire, 50 miles further west. This is the clip used in Wikipedia to illustrate Scottish Standard English.
Which way should we go to Lochwinnoch? One way is seven miles: the other isn’t quite so far, but I don’t want to take the car on that bad road again.
By contrast, here is the voice of the comedian Brian Limond, from Glasgow in the west of Scotland, in a voice that is part‐way towards the Broad Scots end of the Scottish English language continuum.
(Click this text to hear the clip. A transcript will appear here after the clip has played once.)
He says, “Where are we off to, son?” I says, “Just take us intae the toon.” He says, “What wey do you want to go? Do you want to go through Pollokshaws Road, or do you want to go thr‑ ontae the motorway?” I says, “Ontae the motorway”. Driver blahalla, so, “What’ve you been up to [the] night?” and I’m, blahalla, “No much, mate: what aboot yersel?” You know, “What time do you get aff the night?” and that.
“intae” = “into”
“toon” = “town”
“wey” = “way”
“ontae” = “onto”
blahalla, which I wasn’t familiar with, evidently means “patter”.
“the night” = “tonight”
“no much” = “not much”
“mate” (British informal) = “friend”
“aboot” = “about”
“yersel” = “yourself”
“get aff” = “get off”, “finish work”
“and that” = “and that sort of thing”
If you are not attuned to Scottish voices, you may have had difficulty with that. But it is by no means extreme. Brian Limond’s speech is well articulated, as his profession of public entertainer demands. Many Scottish voices, especially from the west of Scotland, are much more impenetrable to outsiders than his.
I intend that readers of this essay should not need any prior knowledge of linguistics. If you are new to linguistics, you may find parts of it to be quite hard work, but I take care to explain everything you need to know. This is an informal essay, not an academic one, and it is not referenced. I avoid technical language wherever possible, but it is not always possible: for example the distinction between a phone and a phoneme is crucial, and I explain it carefully. To indicate pronunciations I use the International Phonetic Alphabet, IPA, and to communicate the meanings of the IPA symbols I arrange that every piece of phonetic transcription – IPA symbols between square brackets – is linked to a sound file. Each sound file is indicated by green text: click the green text to hear the pronunciation. As a taster, here are my pronunciations of “Coke” [kʰokʰ] and “cork” [kʰɒɹkʰ]. And I assure you that Coke makes a better taster than cork.
Scottish English and /r/
Any speaker of English hearing Scottish English for the first time will be immediately struck by how strongly /r/ is pronounced:
I was born and brought up in Torry and I’ve never been out of Torry.
Accents of English can be classified as rhotic or non‐rhotic. A rhotic accent is an accent with post‐vowel /r/: a non‐rhotic accent has no post‐vowel /r/. Most accents of England are non‐rhotic, so that warm is pronounced something like [wɔːm], where the symbol ː after the vowel indicates a lengthening of the vowel. Most accents of the USA are rhotic, so that warm is pronounced something like [wɒɹm] with an alveolar approximant [ɹ]: the tip of the tongue is close to the alveolar ridge behind the top teeth, but not so close as to create a turbulent air flow. Until the 17th century, all accents of English had post‐vowel /r/: so we may say that in rhotic accents this “historic /r/” has been retained, while in non‐rhotic accents it has been lost.
Scottish English is strongly rhotic, and traditional Broad Scots accents are the most strongly rhotic of English accents found anywhere in the world. Scottish English is famous for its “rolled r”, the alveolar trill, denoted [r], as in [ara]. In my childhood the alveolar trill was very frequently heard in Scottish English, right across the language continuum, and in all word positions: pre‐vowel and inter‐vowel as in “three storeys” [θriː ˈstorez], where the symbol ˈ indicates the beginning of a stressed syllable, and even, most spectacularly, post‐vowel as in “door” [dor]. Nowadays it is less prevalent than it was, and it is likely to be heard only in older speakers and towards the Broad Scots end of the continuum, and mostly only in pre‐vowel and inter‐vowel positions. Otherwise it is weakened to an alveolar approximant [ɹ] as in [aɹa] or an alveolar tap [ɾ] as in [aɾa].
In post‐vowel position, an /r/ often creates an extra syllable: “four” [ˈfoɹ̥], where the combining ring below the ɹ denotes a syllabic consonant.
Speakers of Scottish English never have the “intrusive R” of some varieties of Southern British English: “China‑r‑and Japan”. We pronounce ‘r’ assiduously when it is written, and just as assiduously we omit it when it is not written. When a Scot wishes to poke fun at a Southern British accent – not a pastime I would encourage! – the intrusive R is a favourite way to do it.
It is quite difficult, even for a Scot, to pronounce an /r/ strongly in the consonant clusters /rl/, /rm/, and /rn/. Rather than weaken the /r/ in these environments, many Scots continue to pronounce it strongly and insert a vowel after it. Thus “world” [ˈwʌrəɫd], “warm” [ˈwɒrəm], “horn” [ˈhɒrɪn], “pattern” [ˈpʰatʰərən]. As a child I was puzzled by these consonant clusters and I remember asking Dad how I should pronounce them. Around me I heard everything from the Southern British English “warm” [wɔːm] which was ridiculed by my schoolmates as being posh and effete, to the vernacular and (at the time) socially stigmatised [ˈwɒɹam]. Dad encouraged the SSE [wɒɹm] which at first I found quite hard to say. Then when I was 8 we had a family holiday on the Isle of Arran, and I had to be taught that it was [ˈarɪn] and not [ɑɹn].
The upshot is that in normal speech I trill my pre‐vowel and inter‐vowel /r/, but not my post‐vowel ones. The inter‐vowel /r/ in my speech – “Arran” [ˈarən] – may sound to some ears like an alveolar tap, but in fact it is a minimal trill, the tip of my tongue hitting the alveolar ridge twice in rapid succession.
In younger working‐class speech in parts of Scotland, there is evidence that post‐vowel /r/ is beginning to be lost altogether. It will be missed.
Phonetics and phonology, Phones and phonemes
Before I discuss the vowels of Scottish English, I have to explain some subtle but important distinctions between concepts.
Phonetics is the study of the sounds of a language. By contrast phonology is the study of the systematic organisation of the sounds of a language. The difference will become clear shortly.
The sounds of a language are known as its phones. The sounds [ɪ] as in “bit” [bɪtʰ] and [iː] as in “beat” [biːtʰ] are distinct phones. The word gutted is standardly pronounced [ˈgʌtʰɪd] or may be non‐standardly pronounced with a glottal stop [ˈgʌʔɪd], and the sounds [tʰ] and [ʔ] are likewise distinct phones. But – and this is crucial – in the English language the distinction between [ɪ] and [iː] is different in kind from the distinction between [tʰ] and [ʔ]. The distinction between [ɪ] and [iː] has meaning in English: the two phones signify different things. If you say “bitch” in error for “beach” then you have said a different word, and you may risk a slap across the face. By contrast, if you say [ˈgʌʔɪd] in place of [ˈgʌtʰɪd] you have not said a different word: you have merely said the same word in a different way. The contrast between [ɪ] and [iː] is a deep contrast which is part of the systematic organisation of the English language, part of its phonology, whereas the contrast between [tʰ] and [ʔ] is a shallow contrast, a mere phonetic contrast, the sounds being equivalent to each other in the phonology of the language and having the same function.
Phones that are equivalent to each other in that sense are said to represent the same phoneme. Thus in English [tʰ] and [ʔ] represent the same phoneme. By contrast [ɪ] and [iː] represent different phonemes; they differ at the phonemic level. The phoneme that [ɪ] represents is written /ɪ/, and the phoneme that [iː] represents is written /iː/ or /i/. Meanwhile the phoneme that both [tʰ] and [ʔ] represent is written /t/. The phones [tʰ] and [ʔ] are said to be allophones of the phoneme /t/. The pronunciations [ˈgʌtʰɪd] and [ˈgʌʔɪd] are both /ˈgʌtɪd/ at the phonemic level, and they differ only in the surface realisation of the phoneme /t/. Phones are enclosed between square brackets, while phonemes are enclosed between slashes.
A phoneme is therefore an abstraction, but it is not just an abstraction. There is psycholinguistic evidence that phonemes are represented in the mind of the speaker and the listener in a way that mere phones are not.
Formally, a phoneme of a language is an equivalence class of all phones that share the same signifier for the phonology of the language. But I hope I explained it better than that.
The vowel system of Southern British English
It may seem odd to consider the vowel system of Southern British English (SBE) before we consider the vowel system of Scottish English, but there are good reasons.
The vowel system of Scottish English shows big differences from all other varieties of English. These differences are not merely surface differences of how vowel phonemes are realised: they are fundamental phonological differences, differences of structure, differences in the identity of the phonemes themselves. These differences can be usefully shown by contrasting Scottish English with SBE, because SBE, while not at all close to Scottish English phonologically, is in language contact with it and is no farther away from it phonologically than other varieties of English are.
The Southern British English vowel system has 19 vowel phonemes. They consist of six short–long pairs, together with three true diphthongs and four vowels derived from historic /r/. These are:
|/ɪ/ “bit”||/iː/ “beat”|
|/ɛ/ “bet”||/eː/ “bait”|
|/a/ “Sam”||/ɑː/ “psalm”|
|/ʊ/ “pull”||/uː/ “pool”|
|/ʌ/ “butt”||/oː/ “boat”|
|/ɒ/ “cot”||/ɔː/ “caught”|
Vowels derived from historic /r/:
/ɜː/ “sir”, “per”, “purr”
I have not counted the schwa [ə], the weak vowel of the word the [ðə]. The phonemic status of the schwa is problematic, though it may be represented as /ə/ between phonemic brackets for practical purposes.
Unlike phonetic symbols in square brackets, the symbols used for phonemes do not represent particular sounds and so they are to some extent arbitrary. For example the phoneme that I denote as /oː/ has the pronunciations [oː] and [əʊ], amongst others, in different accents – listen carefully to the distinction if you are not very familiar with IPA – and the phoneme itself is denoted /o/, /oː/, /əʊ/ or /oʊ/ by different authors. But it is the same phoneme in all accents – that is, it serves the same function in the structure of the English language in all accents – and so there is no overriding reason to denote it /əʊ/ in the context of SBE just because it happens to be pronounced [əʊ] in SBE. I adopt one convention: there are others.
The vowel system of Scottish English
The basic vowel system of Scottish English has only 12 phonemes instead of 19:
|/ɪ/ “bit”||/i/ “beat”|
|/ɛ/ “bet”||/e/ “bait”|
|/ʌ/ “butt”||/o/ “boat”|
There are four things to notice.
First, notice the heading Monophthongs. A monophthong is a vowel which sounds the same from start to finish. A diphthong changes during its production, as for example [aɪ] changes from [a] to [ɪ]. All the Scottish vowel phonemes in the 2 × 6 table are starkly monophthongal, and the contrast with the long SBE vowel phonemes is clear:
|/eː/ or /e/||day||[deɪ,||deː]|
|/oː/ or /o/||snow||[snəʊ,||snoː]|
|/uː/ or /u/||two||[tʰʊu,||tʰʉː]|
Secondly, the headings Short and Long are absent. In Scottish English all the vowel phonemes are essentially short. That does not mean they are all realised as short in all environments: more of that later. But they are all realised as short in most environments. I lived most of my life not understanding why the vowel phonemes of the English language are said to be long or short: I assumed it was just a conventional labelling that had nothing to do with duration, because all the vowel phonemes have essentially the same short duration in my native Scottish English and I was deaf to the differences of duration in other varieties of English. Conversely, a few months ago I was at a children’s drama rehearsal in Edinburgh in which a young member of the cast who speaks SBE had to say the Scots line, “Mind yer heid!” (Be careful of your head!) But she pronounced the /i/ of “heid” long in her native phonology, and it sounded so silly and wrong. It’s no yer heeed, lassie, it’s yer heid! She couldn’t correct it, and we changed the line. I follow convention by denoting the vowel phonemes that are long in SBE but short in Scottish English as /iː eː uː oː/ in the context of SBE and /i e u o/ in the context of Scottish English, even though they are the same phonemes.
Thirdly, the vowels derived from historic /r/ are missing. It may come as a surprise to speakers of Scottish English to discover just how much it is here that they are freed from. The vowels in Southern British English “eat” and “ear” are quite different: [iːtʰ, ɪə]. It would be an oversimplification to say that the letter ‘r’ is pronounced as [ə]: the initial segments [i:] and [ɪ] also differ, both in quality and in length. Likewise the vowels in SBE “hate” and “hare” are quite different: [heɪtʰ, hɛə], and again the [eɪ] and the [ɛ] differ both in quality and in length. Likewise the vowels in SBE “toot” and “tour” are quite different: [tʰu:tʰ, tʰʊə], and again the [u:] and the [ʊ] differ both in quality and in length. Finally the vowels in SBE “putt” and “purr” are quite different, [pʌtʰ, pɜ:], and again the [ʌ] and the [ɜ:] differ both in quality and in length. In Scottish English, on the other hand, “eat” and “ear” have the same vowel; “hate” and “hare” have the same vowel; “toot” and “tour” have the same vowel; “putt” and “purr” have the same vowel: [itʰ, iɹ; hetʰ, heɹ; tʰʉtʰ, tʰʉɹ; pʰʌtʰ, pʰʌɹ]. By the same token, in Scottish English “sit” and “sir” have the same vowel (for many speakers); “pet” and “per” have the same vowel; “cat” and “car” have the same vowel (for many speakers); “not” and “nor” have the same vowel; “coat” and “core” have the same vowel; “out” and “our” have the same vowel; “Boyd” and “Moir” have the same vowel; and “tie” and “tyre” have the same vowel. [sɪtʰ, sɪɹ; pʰɛtʰ, pʰɛɹ; kʰatʰ, kʰaɹ; nɒtʰ, nɒɹ; kʰotʰ, kʰoɹ; ʌʉtʰ, ʌʉɹ; bɒɪd mɒɪɹ; tʰaɪ, tʰaɪɹ].
The upshot is that Scottish English has no need of any part of SBE’s elaborate system of vowel phonemes derived from historic /r/. In this regard Scottish English is much more simple and logical than SBE. This simplicity may be lost in future if post‐vowel /r/ is lost.
Fourthly, notice the three gaps in the 2 × 6 table. Most Scottish English accents have the Sam–psalm merger, the cot–caught merger and the pull–pool merger, and I will say a little about each.
Sam–psalm merger. Yes, I know it seems extraordinary to speakers of English outside Scotland that here psalm is generally pronounced [sam]. But it is so, and here is the evidence, from Pastor Gordon Mackintosh: “Psalm 1”, [sam wʌn]. And palm is pronounced as “Pam”, and calm as “cam”, and balm as “bam”. The merged phoneme corresponds to SBE /a/ of “Sam”, and the /ɑː/ of SBE “psalm” is the missing phoneme.
Cot–caught merger. Again here is the evidence. “And apparently a couple got caught shagging in there” [gɒʔ kʰɒʔ]. Apologies for the naughty content, but the juxtaposition of the words got and caught was too good an opportunity to miss. Notice that got and caught rhyme exactly, and of course both vowels are equally short: very short in the rapid speech of this particular clip. Generations of Scottish schoolchildren think that 5th November is Guy Fox night, and wouldn’t know whether it is Ralph von Williams and Auto Vaughan Bismarck or something different. The official song of Heart of Midlothian Football Club in Edinburgh includes the couplet,
Our forwards can score and it’s no idle talk;
Our defence is as strong as the old castle rock.
and again as you can hear from this recording, the rhyme is perfect. The merged phoneme corresponds to SBE /ɒ/ of “cot”, and the /ɔː/ of SBE “caught” is the missing phoneme.
Pull–pool merger. Scots share this merger with Pittsburgh in the USA. This time the boot is on the other foot: many speakers of Scottish English cannot believe that these two words are pronounced differently in most parts of England. So, for speakers of Scottish English, here is the evidence, straight from Oxford Dictionaries Online: “pull” /pʊl/, “pool” /puːl/. Again I spent most of my life not knowing this, despite hearing speakers from England around me all the time. The merged phoneme corresponds to SBE /uː/ of “pool”, and the /ʊ/ of SBE “pull” is the missing phoneme.
Vowels in unstressed syllables
In the English language a syllable may have primary stress or secondary stress, or may be unstressed. In an unstressed syllable the quality of the vowel may “reduce” to a “weak vowel”. For example notice the difference in the pronunciation of the letter ‘a’ as between the words “duality” and “dual”. The second syllable of “duality” has primary stress and its vowel is /a/: /ˌdjuːˈalɪtɪ/, whereas the second syllable of “dual” is unstressed and its vowel reduces to a schwa: /ˈdjuːəl/. (The symbol ˌ indicates secondary stress.) In Southern British English, vowels in unstressed syllables always reduce. The 6 short vowel phonemes of Southern British English reduce in unstressed syllables to just 3 weak vowels: the vowels [ə], [ɪ], [ʊ], as in the, in, to when spoken without stress: [ðə, ɪn, tʊ]. (In SBE a syllable with a long vowel or a diphthong always has at least secondary stress and the vowel does not reduce.) In Scottish English, vowels reduce to a much lesser extent. Many speakers of Scottish English, myself included, have the full set of 9 Scottish monophthongs even in completely unstressed syllables, as illustrated by the unstressed first syllables of the following words: “intense”, “except”, “absurd”, “Dundee”, “obtuse”, “react”, “chaotic”, “prudential”, “coerce”. For those speakers, these 9 vowels, at the beginning or end of a content word in an unstressed syllable, may all reduce somewhat, but not as far as in SBE, and none of them to a schwa. In Scottish English the only words that can begin or end with a schwa are function words like “the” and “at”.
I am writing this with my television on, and I have just heard the word drama spoken by a Scot. The Oxford English Dictionary gives /ˈdrɑːmə/ whereas the Scot said /ˈdrama/ with its two vowels identical in quality. In Scottish speech the second vowel of “drama” can reduce to [ʌ] but no farther.
With its full range of vowels in unstressed syllables, Scottish English avoids what has been called the “derdi‐derdiness” of Southern British English: the monotony of one nondescript vowel after another.
I should add that the facts of vowel reduction and stress in SBE are messy, and there are alternative analyses to the one given here.
The “Scottish Vowel Length Rule”
We have seen that all Scottish vowel phonemes are basically short. Certain vowel phonemes however are lengthened in certain environments. The phonemes, the environments and the amount of lengthening vary from region to region within Scotland. Some authors talk of a “Scottish vowel length rule”, but the “rule” is stated differently everywhere you read it. At the very least however, in stressed syllables the vowel phonemes /i/, /u/ and /aɪ/ are lengthened before voiced fricatives (you’ll see in a moment which consonants these are) and at the end of a morpheme (a unit of meaning, which may be a word or a part of a word). Hence short “teeth” but long “teethe”; short “proof” but long “prove”; short “rice” but long “rise”; short “cushion” but long “fusion”; short “side” but long “sigh” and hence long “sighed”. [tʰiθ, tʰiːð; prʉf, prʉːv; rʌis, raɪz; ˈkʉʃən, ˈfjʉːʒən; sʌid, saɪ, saɪ+d]. (The symbol + denotes a morpheme boundary.)
Variation within Scottish English: /ɑː/ and /ɔː/
Not all speakers of Scottish English have all three mergers Sam–psalm, cot–caught and pull–pool. Some SSE speakers, especially middle‐class speakers from Edinburgh and Glasgow, have the contrasting phonemes /a/ as in “Sam” and /ɑː/ as in “psalm”. A smaller number have the contrasting phonemes /ɒ/ as in “cot” and /ɔː/ as in “caught”. Very few have the separate /ʊ/ and /uː/ phonemes of SBE “pull” and “pool” respectively. Where such speakers have /ɑː/ or /ɔː/ or both, they have them as borrowings from SBE, and they tend to borrow them complete with their SBE phonology and a long duration. Also, as befits borrowings, the incidence of these phonemes in Scottish English does not always match their incidence in SBE. Many SSE speakers who say “father” and “rather” with the borrowed /ɑː/ vowel will also say “gather” and “salmon” with the /ɑː/ vowel, even though in English English these are /ˈɡaðə/ and /ˈsamən/.
In my own speech I have separate “Sam” and “psalm” vowel phonemes, though phonetically they are rather close together: /sam, sɑːm/. I have separate “cot” and “caught” vowel phonemes in formal speech: /kɒt, kɔːt/; but not in informal speech: “I caught the ball” /aɪ kɒt ðə bɒl/. It should be clear from what I have said above that I have not got separate “pull” and “pool” vowel phonemes.
Variation within Scottish English: realisation of monophthongs
Within Scottish English there is huge variation in the realisation of some of the monophthong vowel phonemes. We hear “drink” [dɹɪŋkʰ, dɹæŋkʰ]; “hand” [hænd, hɑ̈nd]; “cross” [kʰrɒs, kʰrɔs]; “move” [muv, mʏ:v]. There is considerable variation even within SSE. Although the /ɒ/ vowel of cross can be realised as [ɔ] or even a very short [o], it never merges with the /o/ phoneme or (if the speaker has it) the /ɔː/ phoneme. My own monophthongs are close to the SSE standards: “beat”, “bait”, “bit”, “bet”, “bat”, “but”, “bot”, “boat”, “boot” [bitʰ, betʰ, bɪtʰ, bɛtʰ, batʰ, bʌtʰ, bɒtʰ, botʰ, bʉtʰ].
I assure you that the second voice in the last paragraph above said “drink” and not “drank”. The accent is traditional Buckie, in the north‐east of Scotland.
Variation within Scottish English: /ɪr/, /ɛr/ and /ʌr/
The Scottish English pronunciation of /ɪr/, /ɛr/ and /ʌr/, in words like sir, per and purr, is not stable. Many speakers of Scottish English, myself included, contrast these three endings as I indicated above, but many collapse the three vowel phonemes in these endings to two phonemes or one, most commonly to the single phoneme /ʌ/, so that for example “herd” becomes not /hɛrd/ but /hʌrd/. The collapsed vowel phoneme may differ from speaker to speaker even within the same geographical region, social class, age and gender. Strangely, some middle‐class SSE speakers in Edinburgh and Glasgow adopt as the collapsed vowel the /ɜː/ of Southern British English “sir”, “per” and “purr”, as with this Edinburgh voice:
It’s just more going off and seeing a lot of the same people you’d seen at nursery.
To my ears the Southern British English phonology of [nɜːsʋɪ] sounds strangely incongruous with the Scottish accent of the rest of the sentence. But it is quite a common mixture.
Variation within Scottish English: /ɒr/ and /or/
In SBE, corn and worn rhyme: /kɔːn, wɔːn/. In Scottish English, they do not rhyme: /kɒrn, worn/. Listen carefully to the difference if it is new to you. Thus Scottish English does not have the “horse–hoarse merger” that most varieties of English have: in Scottish English “horse” is /hɒrs/ while “hoarse” is /hors/. But in Scottish English the distribution of the /ɒ/ and /o/ phonemes before /r/ is unstable. For example some Scots have “torn” /tɒrn/, possibly influenced by the similarity of the spelling to words like corn and born, whereas others, even from the same locality, have “torn” /torn/ as though it were spelt ‘toarn’, possibly influenced by the preterite tense tore /tor/. This is not merely a matter of different surface realisations of the same phoneme, like “cross” /krɒs/ [krɒs, krɔs]. Scots who say /tɒrn/ and Scots who say /torn/ are choosing different phonemes: those who say /tɒrn/ are rhyming torn with “corn” and not with “worn”, while those who say /torn/ are rhyming torn with “worn” and not with “corn”.
Phonetically, /ɒr/ and /or/ can be very close together. Here for example is the Scottish author Ian Rankin saying “course”, “force”, “horse” and “Morse” all in the same 30‑minute interview:
That, of course, is problematic.
He said, “I wish I had one like Rebus on the force.”
The Wooden Horse
There’s Inspector Morse
My Scottish ears tell me that he is employing different vowel phonemes, /kors, fors/ but /hɒrs, mɒrs/, though I might have difficulty conveying that to anyone who is not attuned to Scottish speech.
Lexical incidence of vowel phonemes
I have answered the question “What vowel phonemes do Scots have?” and I have answered the question “How do Scots say these phonemes?” I now come to a third question: “What words do Scots put these phonemes in?”
Generally speaking, and subject to the mergers, the complications of historic /r/, and the lack of vowel reduction that we have already seen, any given word will have the same vowel phonemes in SSE as it has in SBE. The word coat may be [kʰəʊtʰ] in SBE and [kʰotʰ] in SSE, but these are both surface realisations of the same underlying phonemic abstraction /kot/. However when we move along the continuum from SSE towards Broad Scots the phonemes themselves may change. Many Standard English words have Broad Scots variants with different vowel phonemes, and many of these words are high‐frequency words. In writing, these phonemic differences may be reflected in a difference in spelling.
Here is one group of examples. Standard English, including SSE, has the words now, down, town, brown, out, about, doubt with the vowel phoneme /aʊ/. In Broad Scots these words are noo, doon, toon, broon, oot, aboot, doot with the vowel phoneme /u/.
Here is another group. Standard English, including SSE, has the words stone, floor, sore, more, most, home, both, whole, and in SSE these all have the vowel phoneme /o/. In Broad Scots these are stane, flair, sair, mair, maist, hame, baith, hale with the vowel phoneme /e/. (The Standard English word hale meaning “healthy” is closely related.) And there are many other such groups. The phoneme changes in such words commonly vary from region to region within Scotland.
These phoneme changes can be confusing, even for Scots. My brother recounts hearing one very proper Scottish lady, trying hard to speak Standard English, hypercorrecting “hailstanes” (hailstones) into “whole‐stones”.
There are thousands of Scots dialect words that are not mere phonological variants of Standard English words, but they are outwith the scope of this essay. (You may have spotted one of them in this very paragraph.)
In this essay I have already said quite a lot about the consonant /r/. Apart from /r/, I have much less to say about consonants than I have said about vowels. For the most part the consonants other than /r/ in Scottish English are pretty much the same as in other varieties of English. I shall just say a little about /x/, /ʍ/, /t/ and /l/.
The phoneme /x/
Scottish English has a /x/ phoneme, a “voiceless velar fricative”, occurring most famously in the word loch /lɒx/ (meaning lake). Most other varieties of English have not got this phoneme. It occurs in a number of other Scottish English words of Scots or Gaelic origin, such as dreich (damp and dismal), quaich (a shallow drinking cup with two handles, often used as a trophy), pibroch (a form of music for Scottish bagpipes): /drix, kwex, ˈpibrɒx/. It occurs in many Scottish place names, such as Brechin, Kirkintilloch, and, comically to non‐Scottish ears, Auchtermuchty: /ˈbrixɪn, ˌkɪrkɪnˈtɪləx, ˌɒxtərˈmʌxtɪ/. Most speakers of Scottish English naturally use /x/ when importing proper names from other languages that have a similar phoneme: “Bach”, “Munich”, “Utrecht” /bɑːx, ˈmjunɪx, ˈjuˌtrɛxt/. In an echo of the classical education for which Scotland was once famous, some speakers of Scottish English use /x/ where ‘ch’ in an English word corresponds to the Greek letter χ in etymology: “technical”, “patriarch”, “arachnid” /ˈtɛxnɪˌkəl, ˈpetrɪˌarx, aˈraxnɪd/>. Many Standard English words with silent ‘gh’ have Broad Scots variants with ‘ch’ pronounced /x/: brought, thought, night, light translating to brocht, thocht, nicht, licht /brɒxt, θɒxt, nɪxt, lɪxt/.
The phoneme /ʍ/
Scottish English distinguishes between “wine” [wʌin] and “whine” [ʍʌin]: it has not got the “wine–whine merger” of most varieties of English. The phone [w] is voiced (the vocal folds vibrate) and [ʍ] is unvoiced. In Scottish English the [ʍ] phone is typically used in words spelt with ‘wh’.
Linguists are divided as to whether Scottish English has an extra phoneme, /ʍ/, represented by the [ʍ] phone. According to one influential school of thought, [ʍ] is best analysed as a mere surface realisation of the phoneme sequence /hw/. In this essay I take the view that there is a /ʍ/ phoneme in Scottish English.
If you need some light relief at this point, I don’t blame you. Spend 30 seconds watching this clip from the Cool Whip episode of Family Guy. Stewie Griffin speaks in an upper‐class English accent that shares the /ʍ/ phoneme with Scottish English.
The phoneme /t/
Inter‐vowel and post‐vowel /t/ are very commonly realised as a glottal stop [ʔ], so that for example “peanut butter” is [ˈpinʌʔ ˈbʌʔʌʴ]. Inter‐vowel [ʔ] is often the community norm in the central belt that includes Glasgow and Edinburgh. “We could even say obligatory for working‐class adolescents”, remarks Jane Stuart‑Smith, Professor of English Language at the University of Glasgow.
Despite its prevalence, the glottal stop is not regarded as Scottish Standard English. But it is not traditional Broad Scots either. It is perhaps best seen as a vernacular departure from the traditional Broad Scots ↔ SSE continuum. It is socially disfavoured, but less so than it used to be.
The phoneme /l/
In most accents of English, the phoneme /l/ is light, [l], before a vowel – ‘lap’ [lapʰ] – and dark or velarised, [ɫ], after a vowel – ‘pal’ [pʰaɫ]. In much of Scotland it is characteristically dark in all positions:
Limmy’s World of Glasgow!
The consonant clusters /lm/ and /ln/ often have an inserted vowel, just as /rm/ and /rn/ can have: “At the scary bit in the film” [ˈfɪɫam].
In post‐vowel position, /l/ can create an extra syllable, just as /r/ can. Thus ‘feel’ [ˈfiɫ̥]. With some speakers, in post‐vowel /l/ the tongue may not even touch the alveolar ridge at all, so that the sound technically becomes a vowel.
My Aunt Elsie, Edinburgh‐born, was minister of a church in Selby, Yorkshire, in the north of England. She had been there for some years, and you might expect that her congregation would have become used to her Scottish accent. One Sunday she intimated from the pulpit that a forthcoming social meeting had a guest speaker, whose subject was “Water plants”. Several members remarked to her afterwards what a strange title that was for a talk. It transpired that the whole congregation, one and all, thought she had said “What are plants?”
If you have followed this essay, you will have glimpsed all the features of the Scottish accent that conspired to create this misunderstanding. From the Scottish perspective, Elsie couldn’t possibly have been saying “what”, because that would have been /ʍ/ and not /w/, but the Yorkshire congregation were deaf to the distinction. From the Yorkshire perspective, it seemed that she couldn’t possibly have been saying “water”, because in their phonology that would have had the long /ɔː/ vowel of SBE “caught”, while Elsie gave it the short /ɒ/ vowel of “cot”. The rhyme of the second syllable of water – the part of the syllable after the /t/ – would be reduced by a Yorkshire voice to practically nothing, and so Elsie’s careful articulation of that rhyme, with its substantially unreduced vowel probably approximating to the [ʌ] of Scottish “purr” [pʰʌɹ] and its rhotic /r/, must have sounded so exaggerated to the non‐rhotic Yorkshire ears that they heard it as a separate word, are. A geographical dislocation of just 200 miles triggered four basic phonological differences in the space of just two syllables.
So now let me recharge my glass with a Coca‐Cola, after carefully removing the cork, and raise a toast to Scottish English:
“Here’s tae us: wha’s like us?”