The American compound rise?

« previous post | next post »

Yesterday, in answering a question from a reader, I glanced over the section on intonation in the 1877 edition of Henry Sweet's "A Handbook of Phonetics". I found what I was looking for, namely the section where Sweet distinguishes three "primary 'forms' or 'inflections' of tones" in the intonation of English — level, rising, and falling — and the "compound tones" such as "compound rising" (= fall+rise) and "compound falling" (= rise+fall).

But next came something surprising:

280.  The use of tone varies greatly in different languages. In English the tones express various logical and emotional modifications, such as surprise, uncertainty, &c. In some languages there is a tendency to employ one predominant tone without much regard to its meaning. Thus in Scotch the rising tone is often employed monotonously, not only in questions but also in answers and statements of facts. In Glasgow Scotch the falling tone predominates. In American English the compound rise is the characteristic tone. [emphasis added]

The part about rising tone "in Scotch" makes sense — this is presumably what's now called the UNB ("Urban North British") rise. But the claim about Glasgow Scotch is contrary to current observation (see Alan Cruttenden, "Intonational diglossia: a case study of Glasgow", 2007; or the examples a couple of LL posts "The phonetics of uptalk" and "Word (in)constancy"). And what in the world could Sweet mean by the claim that "In American English the compound rise [i.e. fall+rise] is the characteristic tone"?

Sweet was generally an acute observer, and not given to making things up, so I'm assuming that these opinions were incautious generalizations of some limited experience. For example, he may have known some Glaswegians who used (at least in his presence) the southern intonations that Cruttenden describes in the paper cited above. But what about the American fall+rise?  Was this some sort of late-19th-century proto-uptalk?

As far as I know, Sweet never visited the U.S., so he must have acquired this opinion from Americans abroad. I wonder who they were, and why they talked like that.


  1. Brian said,

    November 24, 2008 @ 9:03 am

    Hearing the description of the "compound rise", it sounds like it might easily be mistakenly identified in cases where emphasis is on some word in the middle of the sentence– the tone would build up to it and then descend during the sentence's denouement.

    The best example I can think of is the if-then sentence: the speaker has a bit of rise during the "if" portion, and lets the "then" portion fall a bit. Perhaps Sweet's "subjects" were fond of adding conditionals to their statements?

  2. Jubal Musa said,

    November 24, 2008 @ 10:07 am

    "Thus in Scotch the rising tone is often employed monotonously" am I showing my ignorance in thinking this is a wonderful oxymoron? Is there some technical use of the term monotonous of which I am unaware?

    [(myl) It's not a technical use, but the second sense in the American Heritage Dictionary's entry for monotonous is "2. Tediously repetitious or lacking in variety. See synonyms at boring." ]

  3. Tom Recht said,

    November 24, 2008 @ 2:42 pm

    Is it possible that in talking about Glasgow falling tone Sweet was referring only to the tail, i.e. the drop-off after the stressed syllable? From my (limited) experience of Glasgow speech I can see how there's a sense in which you might say it has "falling intonation" – though in fact what makes this end-of-utterance drop especially noticeable is exactly the rising tone on a preceding stressed syllable.

  4. Nathan Myers said,

    November 24, 2008 @ 3:01 pm

    Isn't it necessary, when speaking of tones, to distinguish tones varying over the course of a sentence from those that only affect a word? It seems to me that they may be nearly independent. I can imagine a sentence-level tone variation concentrated in the last word making it hard to distinguish a word tone there, but people seem good at finding ways to deal with that sort of thing.

    I have only heard of word-level tones in Asian languages. Does sentence-level tone variation mean anything in common Asian languages?

  5. Bryn LaFollette said,

    November 24, 2008 @ 3:14 pm

    The description of compound rising tone seems impressionistically like it could describe some sort of stereotyped Southern or Western dialect of the US in 1877, or maybe some stereotyped, generalized "American" dialect. In British English books from around that time period and later, I've often found it interesting that they generalized that the American pronunciation of "pretty", for example, was more like "purdy". Now, I'm sure this pronunciation existed, but I wouldn't think it could really have been typical of the whole of the US. It's possible, I think, that he may have been similarly generalizing the claim about intonation based on either hearsay or stereotypes and a limited personal experience.

  6. Bryn LaFollette said,

    November 24, 2008 @ 3:37 pm

    @ Nathan Myers

    Generally, there is a distinction made between 'intonation' and 'tone'. The sort of "word-level" tones you're thinking of are what is known as phonemic tone or lexical tone, and they are generally wholly seperate from intonation, being associated with the word as much as its vowels or consonants. Intonation, on the other hand, is not to my knowledge ever associated inseperably from the actual lexical content of the phrase it is associated with. Rather, it is a discourse element that conveys information about the speaker's intended interpretation of the associated phrase in the context of the conversation. There are in a given language a sort of stock set of intonational patterns for various phrase types such as questions, statements, commands, etc, and it's the intonational variation between dialects of English that Mr. Sweet is discussing.

    Most every language I'm familiar with uses intonation regardless of whether it uses lexical tones. In Mandarin Chinese, for example, there is sentence level intonation and it serves to convey similar information as intonation in English and other languages which do not have phonemic tones.

  7. Bill Ward said,

    November 24, 2008 @ 8:11 pm

    I was raised (in America) to regard "Scotch" as an offensive word, and that one should say "Scottish" instead, just as one should say "Welsh" instead of "Welch." Is that not the case in the UK?

    But leaving that aside, I too don't know what they mean by an American compound rise. However I'd like to mention an interesting usage that is common among Indian English speakers (those from the Asian subcontinent, not Native Americans)… the use of a rising inflection followed by a pause on the word (usually an article) just before the key word of the sentence.

  8. Nathan Myers said,

    November 24, 2008 @ 8:48 pm

    Bryn: All is revealed. Thank you.

  9. Steve said,

    November 25, 2008 @ 9:22 am

    @ Bill Ward – you're right, in the UK – as elsewhere – 'Scotch' is whisky, (and a few other fixed collocations, like 'Scotch egg' and 'Scotch mist') while 'Scottish' is the nationality. However, this was not the case in 1877.

  10. marie-lucie said,

    November 25, 2008 @ 10:59 am

    … an interesting usage that is common among Indian English speakers (those from the Asian subcontinent, not Native Americans)… the use of a rising inflection followed by a pause on the word (usually an article) just before the key word of the sentence.

    Could it derive from a habit of schoolteachers testing students' knowledge or memory with just such an intonation: "it's aaaa ….. RABBIT!" (or whatever). Students from another language background, taught in English-only schools, would have been exposed to frequent use of this intonation and might have concluded this was normal for declarative speech in English.

  11. Aaron Davies said,

    November 25, 2008 @ 12:22 pm

    "Scotch" = "cheap" (allegedly the origin of the adhesive tape's brand name) is practically obsolete, at least in en-US—probably at about the same level of analyzability as "Dutch" = "bad in general" ("Dutch courage", "Dutch treat", "double Dutch", etc.). In any case, I somehow got the impression that “Scots” was the preferred ethnonym/demonym.

    (Parenthetically, I'd like to note the wonderfully gratuitous use of Scots "outwith" in the demonym footnote of the main infobox on Wikipedia's page on Scotland. Someone has to have done that deliberately.)

  12. Tom Vinson said,

    November 25, 2008 @ 12:48 pm

    The Wikipedia discussion page on Scotland has quite a long back-and-forth about this. The majority of commenters seem opposed to using "outwith", but no one's changed it as yet.

  13. David Waugh said,

    November 25, 2008 @ 12:52 pm

    Sweet must have muddled two different things – a) the pitch of a particular syllable and b) the rise or fall of a whole phrase or utterance. The Glasgow accent often has a low pitch on a stressed syllable (low but not falling) whereas the Edinburgh accent may have a high pitch on an accented syllable. But this has nothing to do with the rises and falls of whole sentences. It is easy to distinguish a statement from a question in both accents, for example. An accent which often seems to me to have a monotonously rising pitch on utterances of various types is that of Ulster but perhaps Ulster people see it differently.

  14. language hat said,

    November 25, 2008 @ 2:08 pm

    The Wikipedia discussion page on Scotland has quite a long back-and-forth about this.

    Long and amazingly stupid. A great many Scottish Wikipedians, understandably proud of their ancient dialect/language, do not seem to grasp that "outwith" is not part of Standard English and should not be used in English Wikipedia. And several people seem to think the "correct" English equivalent is "without," which is also stupid.

  15. Robert Cumming said,

    November 25, 2008 @ 4:30 pm

    I was under the impression that Glasgow uptalk is not universal among Glaswegians, and that the relevant other variable is (or at least was) class. Religion could be another one. Both of these variables could surely have biased anybody's sources in the 1870s.

  16. Nicholas Clayton said,

    November 26, 2008 @ 6:03 am

    You can all engage in a little entertaining fieldwork on this one. A Google search on Rab C Nesbitt will lead to some clips of that (in)glorious Glaswegian on YouTube.

RSS feed for comments on this post