Tricky stress

Halloween (or Hallowe'en) is fast approaching and people all over the world are gearing up for the event.

Halloween originated in Europe as a Celtic New Year celebration. In the Celtic calendar, October 31st was Samhain [ˈsaʊ(ə)n, ˈsɑːwɪn], a pagan festival. The Celts believed that the dead returned to possess the living during that night and so they left food on their doorsteps for the good souls and wore costumes to scare off evil ones. Their priests - the druids - led the people out into the forests where they made bonfires and sacrifices to their gods.
Finally, each family took home an ember from the fire in turnip lanterns, in order to start new home fires. The fires warmed their houses throughout the cold winter and kept away evil spirits.
In the Middle Ages, the Roman Catholic Church established November 1st as All Saints' Day.

According to Wikipedia,

[t]he word Halloween is first attested in the 16th century and represents a Scottish variant of the fuller All-Hallows-Even ("evening"), that is, the night before All Hallows Day. Up through the early 20th century, the spelling "Hallowe'en" was frequently used, eliding the "v" and shortening the word. Although the phrase All Hallows is found in Old English (ealra hālgena mæssedæg, mass-day of all saints), All-Hallows-Even is itself not attested until 1556.

The name Halloween is also phonetically interesting. When pronounced in isolation it is ˌhæləʊˈiːn in RP, thus with main stress on the last syllable and secondary stress on the antepenultimate. But in connected speech it often has the pattern ˌhæləʊiːn, thus having greater stress on the antepenultimate rather than on the last. This phenomenon is called stress shift and is very common in English. Consider the following examples:

a) ˌHalloween ˈparties
b) ˌHalloween acˈtivities
c) ˌHalloween ˈcostume
d) ˌHalloween ˈproducts
e) ˌHalloween paˈrades

In all the above phrases native speakers of English naturally tend to switch around the stress levels in the first element, in order to maintain the regular alternation between stronger and weaker syllable typical of English rhythm.
According to Professor Wells (LPD3, 2008, p.784),

[i]n principle, stress shift can apply to any word that has a secondary stress before its primary stress. In practice, though, it is most likely to apply to those which are regularly followed in a phrase by a more strongly stressed word: most adjectives, but only certain nouns. [...] In some cases usage is divided.

Thus, stresses in English seem to be altered according to context: we need to be able to explain why this occurs, but it's a difficult question and one for which we have only partial answers.

Here are some more instances of the same phenomenon:

a) ˌfourteen ˈstudents
b) ˌChinese ˈpeople
c) a ˌfar-ˌreaching ˈchange
d) an ˌantique ˈchair (sometimes an anˌtique ˈchair)
e) the ˌacademic coˈmmunity

Highlands and Lowlands

On leafing through LPD3, I came across the word Highland and I noticed that Professor Wells syllabifies it as ˈhaɪl.ənd. I then checked the term Lowland and saw it was syllabified as ˈləʊ.lənd. So I wondered: why is that? Do native speakers of English really pronounce these words in slightly different ways? Does it depend on the origins of each term?

If you check in Roach et al. (2006)'s CPD, both words appear as -.lənd, which indicates that in the authors' mental lexicon the words Highland and Lowland are probably analysed as a combination of high/low + land. Not so in LPD, in which the syllabic division proposed by John Wells is not evident from the orthography or from the morphology.
The difference between ˈhaɪl.ənd and ˈhaɪ.lənd lies in the l phoneme. Phonetically, in fact, the first l, being in syllable-final pre-vocalic position, is shorter and weaker than the second l, which is longer and stronger being in syllable-initial position.
Although I was aware of the fact that the LPD's syllabification principles are somewhat different from the ones adopted in CPD, I dared ask Professor Wells why he thought that Highland should be syllabified as ˈhaɪl.ənd and Lowland as ˈləʊ.lənd, to which he kindly answered:

This is not based on any theoretical considerations, just on my intuitions on how I say them and how most people seem to say them.
Compare highly.

Fascinating, isn't it?

If you want to know more about Professor Wells's syllabification principles, read here.

Liʔle Briʔain

Last Monday, one of my students sent me an e-mail saying he couldn't place the accent of one of the female characters in the British comedy sketch show Little Britain. Little Britain, which was first broadcast on BBC radio and then turned into a TV show, comprises sketches involving exaggerated parodies of British people from all walks of life in various situations familiar to the British.

One of the female characters in the show is the teenage girl Vicky Pollard. She is intended to be a parody of chavs (=young working-class people who are rude and aggressive, have a low level of education, and wear a certain style of fashionable clothing such as trainers, sportswear, and baseball caps) living in the West Country. She speaks unusually quickly which, together with the gossip she comes up with, often confuses or annoys the person in question. Watch this YouTube video.

In his e-mail, my student explained he thought Vicky had a kind of a Cockney accent, probably due to the frequent use of glottal stops in her (most of the times) inarticulate speech. But no! Vicky speaks with a typical Bristolian accent. Now, it is true that both Cockney and some West Country accents make extensive use of glottal reinforcement and glottal replacement, but Cockney is non-rhotic and West Country accents (thus including Bristolian) are rhotic. In the video this is easily noticeable: Vicky's r sounds more like a retroflex approximant [ɻ] and also has strong vowel colouring (in some way reminiscent of General American).
Another feature typical of the city of Bristol and not found in Cockney is the so-called 'Bristol l'. This is a very close final allophone of ə sounding almost like FOOT - and thus interpreted by non-Bristolian ears as a kind of dark 'l' - in words ending in orthographic a and ia. If you want to know more about Bristol liquids, here's a link to John Wells's blog that you might find very interesting.

For those of you who would like to find out more about Cockney as spoken today in and around London, here's another link to Professor Wells's blog.

As far as RP's use of glottal stops is concerned, glottal replacement is possible before syllabic nasals but not yet accepted before syllabic laterals. In the phrase Little Britain, the phoneme t can be replaced by glottal stop in Britain (and indeed this is the pronunciation many people use in present-day RP) but not in Little, as this would be considered substandard (cf. Cruttenden's (2008) Gimson's Pronunciation of English, p.82). Here's a BBC video for you in which Prime Minister David Cameron pronounces the placename Sutton as ˈsʌʔn̩.
If you want to know more about this fact, read this.