Analysing "errors": My "shame" list

When I first considered writing this blog, I thought of it as an answer to all my quandaries during my lessons. I was thinking of my students first, in fact. But now that I've embarked on this blogging adventure, I have noticed that I find my own processes fit for analysis, and that I can forge a better path for my students if I retrace my steps as a learner. Well. To be honest, I don't think I've ever stopped being a learner, myself.

This is why this blog post is going to address another of my obsessions with pronunciation which is part of my Evernote Moleskine notebook, travelling with me in my mobile and my bag. I humorously, and dramatically, call it....

The SHAME list!

Image credit:"De Paso Arte"

My "shame-on-you!" list includes a number of words that I have mispronounced systematically (or still do, at times!) (Now, I am not going to reveal his name, but there is a friend and great colleague of mine who also keeps one of these..... ;) ). This post, then, will examine some of my versions of these words, and an attempt at an explanation for my mispronunciations.

Warning! The words "error", "mistake", "mispronunciation" will appear a number of times in this post for the purposes of discussing some of the common views on this aspect of the learning process. I try to adopt a positive view towards error, try to be analytical towards it, and that is the spirit of this article.

***

A bit of theory first. There are different categorisations of "mistakes" around, and one of the views on "error" I first found attractive while a student at College was that outlined by Julian Edge (1989), in his book  Mistakes and Correction. In a nutshell, what I particularly liked about it was this division between:

• slips: those "mistakes" made due to lack of proper attention, which, if pointed to, the student can self-correct.
• errors: those mistakes we make which include features that have been systematised by teachers but which have not yet been "internalised" or "appropriated" or "automatised" by us; 
• attempts: those mistakes which involve the student's experimentation with features that have not been taught, or the overapplication of rules into other contexts.

This leads me into another view that considers the source of error. This great article from Macquairie University explains this very neatly, and I would also like to make a point of some of these categories here:

• Transfer from L1 errors: we all know that we tend to interpret foreign sounds from the filter of our L1 phonology, and as I have discussed in previous posts, the closer to L1 we interpret a foreign sound to be, the more likely we will have more difficulty in learning it (Flege, 1987 and others). Many interlanguage mistakes, then, imply the production of an L1 sound for an L2 quality which we consider to be approximate, or just because it is the only quality we have "at our disposal".
• Spelling-to-Sound errors: these can be fully blamed on the irregular spelling of English (especially if the student uses a transparent, phonetic language as L1), and of course, you can then curse (or enjoy!) the history of English! (I've been reading a great book by David Crystal on this, highly recommended!). Students may merely apply an L1 sound to an L2 spelling looking similar to an L1 spelling, or interpret an English spelling he/she knows about into a context where in fact it is not associated to the sound in quesiton: bʌtcher for /ˈbʊtʃə /, for example.
• Hypercorrection errors: this consists in the overapplication of a rule that does not fit into a context the student believes it should apply to. My secondary school students, for example, apply the /ɪd/ ending in contexts with sibilants, through analogy of the third person singular suffix rule: [wotʃid] for /wɒtʃt/.
• Performance errors: this refers to the errors that may be "instilled" by the challenges of a particular task, that is, lack of weakening by some students when reading aloud, or overchunking when producing spontaneous speech, etc.

A note on another error-related issue. The transfer vs developmental debate: This is a very interesting way of looking at errors. As we have seen before, there are many common mistakes that reveal that the student is actively making hypotheses as to what the language "sounds like", and these cannot be truly seen as transfer/interference errors. At times we are so used to hearing some of these, that we fail to see that several mistakes are actually attempts towards a sound that may bring about difficulty. Take the sound /dʒ/, quite rarely at times used in Spanish as an allophone of /ʒ/ (Yes, another debate as to whether /ʒ/ can still be taken to be the phoneme used for <y, ll> spellings in Riverplate Spanish over [ʃ] or the other way round. Long story! No time to draw on this here!). The video of a local celeb, Susana Gimenez, shows different renderings of the word "you" and some (not all, though) have [dʒ]:

(Sorry about this!) Now. When our students say [ʃip] for "jeep", they are not necessarily transferring this from Spanish, as a true instance of transfer would probably be something like [xip]. This pronunciation which makes our students' rendering of "jeep" homophonous with "sheep" reveals that our learners feel this sound is different and approximate it to a quality they may be more familiar with.

A hypercorrection mistake in this respect may be our students' use of [ʒ] for /ʃ/, as when they say "shopping" as [ʒopin], for example. This is becoming a bit of a sociophonetic feature, as I can easily identify some features of the social background of those students who generally pronounce this this way.

OK. Back to my own errors, then. My own "shame list".

***

 The horror! :p  List 1: [eɪ] for /æ/
The words below stayed fossilized in my interlanguage for a long time. They surely responded to my overapplication, hypercorrection of the /eɪ/ for "a" rule:

*eneɪmel, teɪpestry, eɪbbey (!)

for General British /ɪˈnæməl̩, i-,  ə-/, /ˈtæpɪstri, -ə-/, /ˈæbi /
(Longman Pronunciation Dictionary (LPD) (Wells, 2008), English Pronouncing Dictionary (EPD, 2011) edited by Hartman, Roach and Setter )

There was a sort-of famous song by a local rap band, here, called "Abbey Road". I guess I will have to blame them (ha!) for my serious mispronunciation of this placename while in London in 2010 (felt sooo ashamed!):

In the opposite direction, I remember having problems with the word apron, which I produced as [ˈæprən ] instead of /ˈeɪprən /, and [ˈæprɪkɒt ] for /ˈeɪprɪkɒt, -rə- /  .

List 3: Word Stress
Again, I guess some of these errors stem on either the hypercorrection of the Teutonic tendency towards early stress:  ˈcoincide,  inˈnovative or the overapplication of my perception of the French tendency towards late stress: maˈssage, misˈchief, emploˈyee . I could have also "fallen pray" to the influence of American English.
The proper word stress according to the cited versions of LPD and EPD goes:  /ˌkəʊɪnˈsaɪd, -ən- /, /ˈɪnəvətɪv, -veɪ-, ɪˈnəʊvətɪv / (Wells 2008 has an interesting poll on the pronunciation of this word!), /ˈmæsɑːʒ, -ɑːdʒ /, /ˈmɪstʃɪf / (EPD accepts /tʃi:f/) , /emˈplɔɪiː, ɪm-, əm-/. In this latter case, my own version is accepted /ˌemplɔɪˈiː/ but as the last entry in both pronunciation dictionaries.

List 4: [ʌ] for /ʊ/
Another case of hypercorrection in the words mʌstache, pʌssycat  for /məˈstɑːʃ, mu-/, /ˈpʊsikæt/. In the case of "moustache", American English influence may also be "to blame".
Many of my students also struggle with the word bʌtcher for /ˈbʊtʃə / and  cʌshion for | ˈkʊʃn̩ |.

List 5: Miscellaneous
A few other words I find myself at odds with, include: tournament, handkerchief, gauge, drawer, demise.
I was relieved to see in the dictionary that the first two words also accept my versions:  [ˈtɜːn ə mənt], [ˈhæŋkətʃi:f] and I guess my perceived "error" stemmed from making a spelling-to-sound connection. In the case of the second word, I believe it is the spelling that initially led me into an /i:/ sound.
My versions for the last three words: *[gɔ:dʒ][drɔ:ə][dəmi:z] for the words /ɡeɪdʒ/, /drɔː/, /dɪˈmaɪz /. I guess these three are mispronounced due to different reasons. In the case of "drawer", I thought it was a homophone of the "one who draws", so my analogy backfired here. For "gauge", I don't remember finding another <au> spelling being pronounced in this way and its etymology does not help me at all (I should maybe check the Phonology of Old French!?), so my knowledge of spelling-to-sound connections was used logically, I guess. In the case of demise, I guess it would probably have to do with the fact that the "-ise" ending is generally /aɪz/ in verbs, not nouns, where this ending is generally  /i:z/. So, in spite of applying the rule correctly, I have not pronounced it correctly (such is...English!)

Hope that some of these items of my "shame list" have helped you wonder about the source of your own mispronunciations! (At a very high level of obsession, you'll understand!)

***

To err is human, and "native" speakers of English also engage in frequent pronunciation "errors", and these are reported in some very interesting articles. Some views are really extreme and defo beat my own personal passion, let me tell you!

"8 Pronunciation Errors that made the English language what it is today:" - The Guardian . HERE
"100 Most Often Mispronounced Words in English" HERE
"More than 80% British People Can't Speak Properly" HERE
"17 Everyday Words you may just be Mispronouncing" HERE

Any pronunciation pet-peeves out there? Post them on the "comments" box!

Ideas on the go #1: Creating a mental image of tone

At one of the Teacher Training Colleges where I work, most second year trainees in their Laboratory Practice II course (the oral practice component of the Phonetics II subject)  are exposed to one of the most widely-used materials in the world of intonation: Intonation of Colloquial English, by O'Connor and Arnold. In my humble and inexperienced opinion, this classroom material works fine to allow students to become aware of pitch changes and basic English tunes but not to help trainees see how tones create or help us negotiate meaning in interaction.  Mind you, none of the teachers in the staff, as far as I know, considers the "attitudinal" aspect in depth to teach the tunes, as the theoretical framework in the Phonetics courses centers around the phonological contributions of Discourse Intonation and Systemic Functional Linguistics (You can read a great article on these ways of approaching intonation by my colleagues Lucía Rivas and Miriam Germani here) In spite of this, I have to admit I have not yet found a set of materials that so exhaustively provides models for different patterns except, perhaps, for the practice sections in the book by Wells (2006). 

My experience many years after learning intonation via these materials is that the O'Connor and Arnold "experience" has conditioned by ability to hear and make sense of tunes. I have found it so hard to start "re-adapting" my perception of what is really going on in English, and the materials by Jack Windsor Lewis, and Roger Kingdon, however outdated they may seem to some people, have really helped me "break the mould" when it comes to perception of tones. However, I am a phonetics "freak" myself and I do care about phonetic and allotonic detail, but it is true that the native speaker processes the signal and assigns some meaning to it, he/she has some sort of "action-activity recognition" from a Conversation-Analysis perspective, and that is what matters in terms of the perception of intonation, I guess. An probably, this is what the teachers-to-be I am training need, so a collection of materials for oral practice and theoretical analysis of tones like the ones my colleagues have made at College does seem appropriate, as long as we don't lose touch with the real use of language "out there".

***

Anyway. Digression again, as I was not planning to get into an informal assessment of intonation materials (this deserves a better post, and I will write about this some time in the future). Today, I wanted to address a set of difficulties some students presented with the modelling of tones.

During my tutorial periods, students were given a set of drills from O'Connor and Arnold  so as to read the patterns presented in the sentences. In spite of the fact that students do have the audio files of the book at their disposal and are supposed to imitate the audio, and do so pretty systematically, many students find it difficult to apply the patterns to a new, unrecorded, drill. Away from the model, there is no mental image to link the tune to.

At times, the image of the tunes in the students' minds are associated to a drill they remember by heart, and this helps, but at times, they find the transfer to a new context difficult. In this respect, I have tried two tips:

a) For students who need a reference drill:

Inspired by the DAR-DOOBY-DIPETY classic task by Mark Hancock, which I have rediscovered thanks to his presentation at IATEFL (also by the reference to the activity Krystina Poesova has made in her great talk at Harrogate!), I have found one out of many possible ways of helping students get a reference for tones. I have asked these students to record a DAR-DOOBY-DIPETY (and a four-syllabled word) DIPETITY version of each of the tones, in a different key, on their mobile phones. This has helped some students to have a ready-to-grasp voice note, recorded in their own voices, to keep as reference for the production of the tone for a different number of syllables as tails. When faced with a sentence with tone marks to read out, students would compare their versions to the DAR language ones on their mobiles (which they can retrieve really quickly!)

Here's an example of what I have asked my students to record on their mobiles:


I believe that the use of non-words for intonation reference may work, as it keeps meaning associations away from the tunes. In many occasions, students see the content of the words and apply a tune which they feel suits the content, but it does not match the pattern written in the book. (This, of course, makes for a great opportunity to discuss the different contexts that different decisions on intonation may project, but  the purpose of the activity in question is mostly to get students to imitate or produce certain patterns in a more systematic way). I should maybe see, in the future, if getting students to use the same reference words for the tunes may backfire and lead to further confusion ( a creation of non-sense words for each different tone may turn out better), but so far, in the limited number of students I have tried this humble tip on, it appears to have worked!

b) For students that may not be aware of what they are producing or have trouble imitating the "master":

In this case, I have encouraged the creation of a sort of "attitudinal image" in the students' minds. Although I have made a point of the fact that attitudes may not always help them make an appropriate choice of tone, I have asked some students to associate an attitude, feeling or reaction to the tune in question. Some of these ideas were also related to Spanish feelings or expressions. These attitudes may or may not match what the tone is "doing" in that particular context, but the focus of this activity is to build a mental image of what the tone sounds like. So, for example, these are some of the mental attitudes that did the trick with those students who were at first unable to reproduce a tune:

High fall: complete refusal of something, as for Spanish: No! Estás loco!. Also for complete surprise: Copaaado!. (Translations: No! You must be mad! and Wicked!)
Low fall: not surprised, it was obvious it was going to happen, as in Spanish: buah. Also sequence-closing, as for an argument: Se acabó. (Translations: Yep, duh, and It's over) 
Low rise: in general, the idea of counting objects with kids works as a mental image here. In order to make students aware of the step-down to begin the rise, I sometimes ask them to remember the voice of the Addams Familiy's butler, a sort of creaky-voice beginning:

High rise: this I have found a bit more difficult to model, and I generally ask students to go for a more "falsetto", "Shakira-like" note on the last syllable of the intonation phrase, mostly asking them to worry less about the starting point of the rise, and focus more on the end-point (avoiding, of course, an unnecesary prominence in the last syllable of the tail):

Fall-rise: Though not exactly the same in Spanish, many students get to associate the idea of a non-committal, insincere Sí, or Puede ser to this tone, in a context like "do you like my dress? (it looks awful).

From these mental images, we teachers start fine-tuning the productions of tones for each student individually till students find their own way of producing the tones on the page, or the tones expected in a particular task. They will, later, with the support of the theory in their Phonetics courses, make their own choices in the un-marked texts and spontaneous speech tasks carried out in their courses.

(Of course, all these initial notions and searches for the mental images needed to produce tones are fine-tuned individually and gradually. It is by no means our intention to harm students' vocal folds in any way or to force unnatural patterns out of them. These tasks are useful for the initial stages when gaining awareness of vocal range and possible glides is necessary, and continuous feedback and follow-ups on the part of the teacher then ensure that learners find a voice and a range that suits them best.)

I have not discussed the step ups and step downs for the onsets in this post, and for this, like some of my colleagues have done, I use the piano or the harmonica apps on my mobile, and encourage learners to do the same. There are other tips "out there" which surely do the trick, so feel free to share them in the "comments" box below!

Hope you have found these tips useful!