Rhythm and clues

I have dedicated most of my academic life to research on multilingualism, whether simultaneous or consecutive, with a focus on pronunciation and related matters of accent. My take on accent is broader than what I’ve understood this word to mean among specialists as well as laypeople. To me, accent relates not just to vowels and consonants, but principally to prosody, the rhythmical melody which necessarily associates with speech and which, therefore, is necessarily present whenever we speak our language(s).

Calls for attention to the fundamental role that prosody plays in spoken communication are not new. In a previous post, I quoted the earliest references I could find which deal with prosody in child language acquisition. Another interesting study dates from 1946, titled ‘Psychological aspects of speech-melody’, where Louise Zucker pleads for the inclusion of intonation in adult language teaching. Her observations among immigrant populations led her to conclude that newcomers to a foreign country can’t feel “psychologically” at home in it without awareness of the relevance of speech melody.

More recently, in 2010, Reyna Gordon and colleagues investigated the psychological bonds holding words and melodies together, in ‘Words and melody are intertwined in perception of sung words: EEG and behavioral evidence’. They found that “variations in musical features affect word processing in sung language”. This reinforced my conviction that attempting to use “words” without their associated linguistic melodies is like attempting to cook a meal by leaving its ingredients intact. To my mind, findings about sung words are no different from findings about spoken words, because spoken words are sung – whether we choose to call them by this or any other name.

What this research makes clear is not that different languages, or different language varieties, sound different: we all know that. The insights are first, that prosody contains clues to meaning, which we must be able to both listen for and produce; and second, that isolated sounds or words tell us nothing about these clues, because nobody speaks in isolated sounds and words – except in language classrooms. As I’ve repeated over and over again in this blog, starting here, traditional language teaching syllabuses, methodologies, and assessments continue to rely heavily on printed modes of language. Standard orthographies of any language tell us as much about how those languages are spoken as a printed recipe tells us about how food tastes. We don’t speak in ingredients.

Another core reason, in my view, for the neglect of prosody in language teaching is that we have come to assume that learning languages is an intellectual task, whereby we start with conceptual mechanics, to then strain our memory attempting to remember what our new language *looks* like, at the bottom of that right-hand page in the textbook. I’m sure I’ve seen telltale signs of these retrieval exercises in learners’ hand and eye movements, and I’ve had learners confirm that this kind of information is what they’re trying to recall. In short, we’ve forgotten that we use our bodies to speak, and that our bodies therefore harbour fundamental clues to meaning.

Image © Clipart from clipartheaven.com

As early as inside the womb, we’re primed to the rhythms of our body, whether physiological or linguistic, with no visual aids. There is no light inside the womb, but there is sound – and I’ll have a lot more to say about this in my next post. Let me just add here that linguistic rhythms, unlike, say, cardiac ones, aren’t inbuilt. Linguistic prosodies have grammars, which is one reason why speaking a language with disrupted prosody results in disrupted intelligibility, as stand-up comedians know well. Prosodic false friends abound as much as lexical ones, the difference being that we don’t talk about them. When we assume that a particular falling or rising tone, or a particular tone of voice, mean the same across languages, we’re (mis)taking similar function for similar form, just like when we assume that English deception means Portuguese decepção. The same is true within the “same” language, incidentally: my Brazilian friends find my way of singing “our” language endlessly funny, as much as my habit of putting on a camisola when I feel cold, which to me is a sweater and to them is a nightie.

If you’re curious about how we sing as we speak, have a look at these two publications. My article ‘Portuguese and English intonation in contrast’, and Daniel Hirst and Albert Di Cristo’s book, Intonation Systems. A Survey of Twenty Languages, where I also have a chapter, and which gives the first overview of uses of prosody across and within languages.

Next time, as promised, I’ll explain why we’re all natural singers and dancers, and why this is relevant to our linguistic abilities.









Cruz-Ferreira, M. (2004). Portuguese and English intonation in contrast Languages in Contrast, 4 (2), 213-232 DOI: 10.1075/lic.4.2.03cru

Gordon, R., Schön, D., Magne, C., Astésano, C., & Besson, M. (2010). Words and Melody Are Intertwined in Perception of Sung Words: EEG and Behavioral Evidence PLoS ONE, 5 (3) DOI: 10.1371/journal.pone.0009889

Zucker, L. (1946). Psychological Aspects of Speech-Melody The Journal of Social Psychology, 23 (1), 73-128 DOI: 10.1080/00224545.1946.9712314

© MCF 2013

Next post: Rhythm clues and glues. Saturday 4^th May 2013.

Shibboleths & Co.

The word shibboleth originally designates a specific pronunciation of itself. Failure to pronounce the word in one native way, namely, substituting an ‘s’ sound [s] for a ‘sh’ sound [ʃ], entitled [ʃ]-natives to hack [s]-natives to pieces. It could have been the other way around, of course, given our propensity to nurture murderous feelings towards whoever speaks differently. It all depends on who is wielding the righteous axe at any given place and time.

Many of us have come to believe that making mincemeat (literally) of differently-accented people isn’t really an acquired right, so we’ve come up with civilised ways of obliterating such people instead. We can bully them in school, for example, or refuse them jobs, or force them to lose academic visibility, or just poke fun at them. We can also talk about them as belonging to ghettos, which is a bad thing to belong to, for those of us who belong to good ones.

Likewise, shibboleths can be good or bad. It all depends on who is wielding the righteous standard at any given place and time. When you enrol in language courses, you’re destined (I was going to say “doomed”) to learn what your textbooks deem to be good shibboleths. Whether you like them or not, whether you need them or not, for the purposes which made you enrol in those courses in the first place.

Sometime into your language learning, you are likely to be told that you speak your new language “with an accent”, and should therefore strive to get rid of it, on the (extremely entertaining) assumption that speaking a language well means speaking it without an accent. The claim that a good accent means no accent probably stems from the widespread practice of teaching languages through printed media. Printed languages have no accent, so it’s up to whoever happens to be reading them to provide them with one. Just look (yes, “look”) at the different accents which have been attributed to Latin, for example. This also means that whoever is in charge of teaching you a language can always claim that whatever accent you have in it doesn’t match the “good” one intended by the textbook writers, because they’re native speakers and you aren’t, and native speakers can’t be challenged on matters of “good” language usage. That argument usually cows learners into accent submission.

“Good” and “bad” accents, however, are not simply a matter of native vs. non-native pronunciation practices. As with the original shibboleth incidents, native speakers are more than willing to chop figurative heads (and so offending vocal tracts) off one another as well. I’ve blogged about this before, but Richard Cauldwell, in an article titled ‘Lord Rant: A personal journey through prejudice, accent and identity’ puts it much better than I ever could.

Our accents reflect the ways in which we’ve trained our vocal tracts to produce speech. They match not just what we hear around us but also, and importantly, the ways in which we want to be heard. Until around age 3, we’re not aware that we’re acquiring accents in our languages (or the languages themselves), because we’re not aware of ourselves as independent from our surroundings. From then on, we start being able to pick and choose, be it our friends, our clothes, or our dialects, complete with accent, all of which will become part of who we are. We acquire and lose accents for the same reasons that we acquire and lose languages, and we do this at any age. Just listen to one-time fellow-speakers of your dialect(s) who, as adults, moved to different countries, or different regions of the same country.

Being able to adapt to the surroundings of our choice is a condition of survival. That’s why we, as a species, have been able to inhabit (I was going to say “infest”) every nook and cranny of the world that we are able to perceive. We will adapt, including our accents, if we so wish. Not all of us need or want to learn a new language in order to impersonate some of its other speakers. Learners may want to distance themselves from specific accents of their new languages because, well, learners aren’t free from prejudice either. Perhaps learners will want to acquire the shibboleths which identify them as borrowers of a language, rather than the shibboleths associated with specific lenders of that language. This includes prosodic shibboleths, more on which next time.

© MCF 2013

Next post: Rhythm and clues. Saturday 20^th April 2013.