I recently stumbled upon a remarkable 1893 tome on Google Books entitled Some peculiarities of speech in Mississippi by the delightfully-named Hubert Anthony Shands. A glossary of words native to the dialect(s) of that state, the book opens with a detailed description of Missisippi pronunciation habits.
Of course, this being the Victorian era, Shands pens a number of howlers within the first few pages, such as:
I suppose this confusion of (i) with (e) is due to laziness or sheer negligence, as surely the great majority of people must possess ears delicate enough to readily distinguish between the two sounds…
Despite such archaisms, the book is remarkable for what it highlights as well as omits. In some instances, Shands confirms the presence of modern Southern English features in the 19th-Century. For example, note this description of African American English:
‘Th’ (ð) is nearly always pronounced as ‘d’ at the beginning of words, by negroes; as, (dis) for ‘this,’ (dæt) for ‘that,’ (dem) for ‘them,’ (den) for ‘then,’ etc. At the end of words, negroes generally give (ð) the ‘v’ sound; as (smûv) for ‘smooth,’ (brîv) for ‘breathe,’ (sûv) for ‘soothe,’ etc.
This corresponds to contemporary African American Vernacular English, which has preserved this distribution of consonants 118 years later. By contrast, many other features Shands ascribes to African Americans suggest a very different dialect from what can be heard in 2011.
Other passages hint at modern accent features, but suggest striking differences. Is Shands describing the pin-pen merger in this paragraph (the tendency for Southerners to pronounce ‘e’ as an ‘i’ sound before ‘n’ or ‘m’)?
The confusion of [the ‘e’ sound] with (i) is very common indeed among the illiterate classes, and is heard quite often among the educated: (simineri) for ‘seminary,’ (simiteri) for ‘cemetary,’ (sit) for ‘set,’ (pin) for ‘pen,’ (ʧist) for ‘chest,’ (git) for ‘get,’ etc.
Shands asserts that ‘e’ becomes ‘i’ not only before nasal consonants, but before all consonants generally. Was this pronunciation indeed true of all such ‘e’ words in 1893? Or is Shands overgeneralizing? I can’t say for sure.
And then there’s one very conspicuous omission. Contemporary Southern accents are defined by the vowel in words like ‘time,’ ‘fly’ and ‘five:’ these become a monophthong (‘tahm,’ ‘flah,’ and ‘fahv,’ to put it crudely). But Shands suggests something different in 1893:
Long ‘i’ (ai) is nearly always correctly pronounced, and seems to follow no rule in those changes that it does undergo. There is no group of related or similar words in which it suffers regular change.
Love the use of ‘suffers’ there. Anywho, this is noteworthy in light of modern Southern pronunciation. Although as with the previous quote, I can’t draw conclusions.
Of course, there are pronunciations Shands describes that don’t correspond to modern Mississippi accents at all. He finds that the lower classes pronounce ‘garden’ as ‘gyarden’ and ‘car’ as ‘kyar’ (typical of contemporary Jamaican dialects!) ‘Palm’ and ‘psalm’ apparently rhyme with ‘Pam,’ while ‘parcel’ rhymes with ‘tassel.’ ‘Stamp’ and ‘tramp’ rhyme with ‘stomp’ and ‘tromp.’ ‘Boil’ and ‘boy’ rhyme with ‘bile’ and ‘buy,’ much as they do in Irish English. And this is only scratching the surface.
You can browse through the whole text here. I haven’t approached the enormous word list yet, which I’m sure is full of goodies. Have fun reading, and if you note unusual terms or pronunciations, let me know!
Absolute gem of a book by the looks. However, be careful with his statements especially:
Long ‘i’ (ai) is nearly always correctly pronounced, and seems to follow no rule in those changes that it does undergo. There is no group of related or similar words in which it suffers regular change.
Where does he place “correct”? A victorian proto-RP speaker would have had nothing like today’s pronunciation of “time”, and I’m willing to bet a southern gentleman of the post-bellum would be unlikely to have the same concept of what that vowel “should” be than a modern speaker of general American.
Alas, being a product of his time, Shands doesn’t really elaborate on what the ‘correct’ pronunciation is. His use of ‘ai’ would strong suggest a diphthong to me, but who knows? My impression is that he bases his judgments on some standard other than Southern speech, but I haven’t the foggiest idea what that standard is!
I suspect it’s something very different from modern American “ai”, I bet it’s a lot closer to what was being said in Boston or London than what was emerging from the future midwestern heartland of General American. The Mississippi “time” could then have travelled in the opposite direction from the rest of the US, but from a starting point intermediate between today’s standard and today’s broad Mississippi accent.
Actually I don’t think the vowel of “time” has changed much between Victorian proto-RP and contemporary General American.
I find it interesting that Shands regards CLOTH-broadening (words like “off” having the THOUGHT vowel rather than the LOT vowel) as deviant, and the use of CLOTH=LOT as “correct”. (p. 11 (Google Books link)).
I have two reasons to be surprised by this:
* CLOTH-broadening has always been the standard in North America (for those with cot != caught), has it not?
* Even if Shands was using a British prestige standard, I should have thought that the use of THOUGHT in words like “off” would have been standard in the Received Pronounciation of the time.
PronUnciation. Aaaaaarrrrrrrgh!!!
I was quite surprised by his acknowledgement of CLOTH as well. It was standard in RP and probably the vast majority of American accents at the time. I also wonder why he continually pointed out the non-rhoticity of Mississippians: assuming this work was geared toward the London, New York, Boston and Southern aristocratic intelligentsia, would r-lessness be assumed to be the norm?
Another one: he thinks that “herb” ought to have initial /h/ (p. 12, section 19).
British standard, maybe not at the time though.
Of the three words he gives for “confusion of (i) with (e)” that don’t have it before n or m, really only “chest” is unusual.
saying (sit) where I would say (set) surely happens, just as there are people who would sometimes say lie where I would say lay. Indeed “sat it down” gets a number of Google hits. So a “sit” pronunciation, if it happened, was, I think more likely using the word “sit” than pronouncing “set” as “sit”.
And “git” is indeed a current pronunciation of “get”. Don’t know where that comes from, but maybe it relates to whatever it was he was observing.
I pronounce “get” and (sometimes) “again” with KIT, but do not have a pin-pen merger (nor a Southern accent). So those seem to be separate phenomena, although I can readily believe that he heard both in Mississippi. As for lay/lie, sit/set and raise/rise, I think those may vary everywhere.
For me as well, get can be KIT (and also DRESS). I agree that the hearing of set/sit is due to lexical differences in the dialect. The author was likely hearing a completely different verb!
It’s probably an ignorant overassumption like how many believe Canadians merge about with boot. Contrary to what the Canucks say, they do say it very differently from the Yanks, but the diphtong’s onset is only raised similar to how both Yanks and Canucks raise their i in life and light.
The general raising of ‘e’ words may come from Ireland. See A Sound Atlas of Irish Englishby Raymond Hickey. John Wells also makes a very brief mention of this on p. 423 of his Accents of English. The “kyar” and “gyarden” thing can of course still be found in Ulster to this day, even among young people in some areas.
I believe you can find some hint of the pin-pen merger in Western Ireland even today. (Not to mention the traditional pronunciation of ‘devil’ as ‘divil’). Showing that ‘general raising’ came from somewhere would be hard, though, since this vowel shifts upward and downward quite a bit.
Well, I think it’s Irish.
The mergers in “chest” and “get” are motivated by the preceding consonant. I don’t think he was hearing the same vowel in them, though. The vowel of “chest” is, and probably was, more of a [ɨ].
Pingback: This Week’s Language Blog Roundup | Wordnik ~ all the words