Pronunciation and Grammar: The regular simple past inflection -ed

One of the things I try to instill the most on my teacher trainees is that pronunciation can easily be taught alongisde other language areas and skills. I have found this makes many teachers feel relieved and more confident about teaching pronunciation in the EFL classroom, as there are generally many myths and misconceptions regarding the way phonology should be tackled in the lesson (more on this on a future post!).

One of the obvious choices of pronunciation features to teach together with grammar arises when the simple past is taught, specifically the regular past inflections -(e)d.  

The Rules
A brief review of the way the rule goes:

We look at the infinitive form of the verb:

• If the last sound is voiced (except for /d/), we use /d/.
• If the last sound is voiceless, except for /t/, we use /t/.
• If the last sound is /t/ or /d/, we use /ɪd/ (in other accents of English, /əd/ also possible).

This constitutes a great opportunity for students to become familiar with the concepts of "voiced" and "voiceless", to feel the vibration of their vocal folds for voiced sounds by placing their hands on their throats, or covering their ears to feel the "buzz", or sensing the vibration on the top of their heads upon the production of a voiced sound (the contrast /s/ /z/ (the "snake" and the "bee") is generally a great choice!).

Other teachers "shun" the presentation of voiced/voiceless contrasts by teaching this rule by referring to spellings . In my humble opinion, why complicate matters further by engaging the memory in endless lists when in fact voice/voicelessness can be "felt"!

A few warnings!

I feel it is my "moral duty" to make a few observations regarding the pronunciation of regular past tense endings and their teaching:

1. We cannot assume students will know how to pronounce all the infinitive forms of the  verbs they encounter, and they may make mistakes. For example, a Spanish speaking student not yet familiar with the sound /dʒ/ may deem the last sound in the verb "judge" to be a /ʃ/ or a / tʃ/, for instance. So this will result in the wrong choice of -ed pronunciation.

BUT, more importantly....

2. The contrast /t/-/d/, to be honest, is not always kept. Why? Well, simply because in final position before silence, or when followed by a voiceless sound, /d/ will present varying degrees of devoicing [d̥], that is, it will lose vocal fold vibration during the whole or part of its production, and will thus approximate /t/ in its production. There will also be no audible release if followed by  affricates or other plosives, or by silence (optional, according to Tench 2011:66) , or subject to delayed release if a homorganic (a consonant sharing the place of articulation, in this case, alveolar), follows. That is, the final stage of the /t, d/ will not be heard, you will be keeping the release "to yourself", so to speak, till the next plosive is produced. A more technical discussion of this can be found in this blog post by John Wells. Another very interesting and illustrated discussion of these allophones can be read about in Knight (2012).

So what should I be focussing on as a teacher? Primarily, vowel length. Vowels preceding the ending /d/ will be longer; compare: complained [kəmpleːɪnd], complaint [kəmpleˑɪnt]. It has been established by both ELT and ELF pronunciation experts that vowel length may be an important source of confusion and unintelligibility at times, so it is something worth teaching at some point or another.

You could also work on the avoidance of an "exaggerated" ending to work towards a non-audibly released /t/ or /d/ in the contexts where release is not usually heard. However, I know many teachers who would prefer an exaggerated production of the ending as "proof" that the past tense is being produced. And I have to say, it may pay off in oral Language and Grammar exams! Better to be heard producing the past than having to explain (if you are ever given the chance!) that in fact you know your Phonetics theory well!

BUT, even "worse"....

3. When these past forms are followed by other words starting in consonants (except perhaps /h/) resulting in a cluster of three consonants  in which /t/ or /d/ are medial, these past markers can, and sometimes are, elided, as Tench (2011:96), explains. Cruttenden (2014:314) describes the tendency of elision of voiceless continuant + /t/ or voiced continuants + /d/ followed by a consonant at word boundary: finished late /fɪnɪʃt leɪt/ , refused both /rɪfju:zd bəʊθ/. Similarly, elision will apply in combinations of voiceless plosive/affricate + /d/ or voiced plosive/affricate + /t/: , looked like /lʊkt laɪk/, rubbed gently /rʌbd dʒentli/. Cruttenden claims that it is the context that makes up for the lack of the past tense marker in these cases.

Jack Windsor Lewis, however, claims they are generally not elided if followed by an obstruent or nasal (#11 - here) and focuses on the producing of the right non-audibly released allophones instead.

So what do we do?

If we want our teaching of English to be "enabling" and "empowering" for our learners, we need to teach the rule as it is. We may focus on the contrast /t/ - /d/ to different degrees depending on the objectives our class needs to meet, but I personally believe there should be at least a contrast between some sort of [t] sound (be it an attempt to a devoiced [d̥] or other "version" of /t,d/) vs /ɪd/. Students' rendering of past forms often include an overgeneralization of the rule, with varying versions of the /ɪd/ ending produced for all verbs, [wotʃed, wotʃid] even [ɣwotʃid]  for "watched", to mention a very common example. So I claim that there should be at least a contrast between the /ɪd/ ending and some other variant for /t, d/, which you will define according to your students' needs and level.

Truth is, with this basic two or three-sided contrast, we have already got a lot on our plate, as the production of consonant clusters is a huge challenge in itself, and at times, students appear not to produce their past tenses in speech because of the cluster rather than because of lack of knowledge of the past tense in itself, as my colleague Prof Terluk rightly points out.

How do we assess this?

Lists of regular verbs can be used for both perception and production, and different experiments can be carried out by students to check whether the last sound of the infinitive is voiced or not. In my experience, teenagers find this exercise pretty entertaining (Just don't "spoil" it, as I did, by showing videos of the vocal folds in operation....couldn't help myself, I guess!).

The web is also filled with automatic-grading tasks for students to try and identify the right ending (check the collection of resources below) to extend pronunciation work beyond the classroom.

Whenever a written exercise on the simple past tense is checked in class, students may be asked to also pay special attention to their oral rendering of the past verbs. My secondary school students would deem an answer given by a classmate as "wrong" if the past tense was not pronounced correctly, even if the written version was correct. Thus, you can always make the most of a "boring" grammar exercise check to practise pronunciation! (Killing two birds with one stone, that is!)

For a successful grammar-pronunciation integration to take place, the production of the 
-ed ending needs to be monitored and, yes, assessed. A very successful task I tried on my secondary students consisted in their planning and recording a criminal confession (the unit was related to "Crime") in which at least 10 regular past verbs had to be used. They would then have to self-assess their production of past tenses by counting the number of regular past verbs and the number of -ed endings produced correctly in their recording, to give themselves a percentage of correct answers. These percentages would then be "double-checked" by a classmate. This helped me realise that at times students are harder on themselves than we teachers can be with them, feedback-wise!

Resources!

As usual, here's my share of collected resources to practise the application of the -ed rule:  http://www.scoop.it/t/pronunciation-bites/?tag=ed+ending
Enjoy!

***

Hope you have found this post useful, and as usual sorry about having to reveal the "dark side" of a topic which, at first glance, appears to be easy to teach. When you've given it enough thought, you will notice it is indeed simple to teach, and you will, of course, make your own choices and adaptations based on your own groups and courses. Good luck!

Teaching Segments #1: English /ð, d/ - Part 2: Phonological Processes and "Tips & Tricks"

In one of our previous posts, we talked about the allophonic spit Spanish learners are faced with, from  [ð, d̪] to English /ð, d/. That is, in English, /ð/ and /d/ are two separate sounds, in Spanish, they are two variables of the only phoneme /d/, with some differences in place of articulation as well.

Interlanguage Processes in the Production of /ð, d/ for advanced learners of English with Riverplate Spanish as L1

Let's take the following phrases from the dialogue "At a Party", from Anne Baker (2006) Ship or Sheep, 3rd edition:

In the dark!
They are dancing under the stars!

The Contrastive Analysis hypothesis by Weinreich (1953) and Lado (1957) would establish that after a thorough comparison of phonemic and phonetic inventories of both L1 and L2 that defines what sounds and variants are present or missing in both languages, we can predict that my learners would apply their L1 allophones Spanish [d̪] and [ð], thus:
(Assuming that learners know about the non-rhoticity of GB)

In ðe ðark!
d̪ey are ðancing unðer d̪e stars!

However, we cannot just assume that students, especially if they are advanced learners (B2, C1 level of the CEFR) will just transfer their L1 sounds to the L2.  They may have motivations to do so, and not just because the foreign sounds are not present in their L1, but also, because they may perceive them to be similar. The presence of "inaccurate perceptual targets" is described in Major’s
Similarity/Dissimilarity Hypothesis (2006), which  contends that those features of a language which are perceived as similar will be more difficult to acquire, since they are perceived and interpreted in terms of L1. As established earlier, Spanish /d/ may be heard to be similar to English /d/, as they are both plosives in nature, and the dental and alveolar articulations are not really far away from each other, and something similar may happen to the dental fricative, there being one allophonic variant of /d/ in Spanish under that description. This hypothesis may also explain why other sounds, such as /dʒ/, non-existing in Spanish and only closely associated to /ʃ/ or /tʃ/ by Spanish speaking learners, may be more easily acquired or "approximated to" and sustained if students' attention is drawn to it. (I am not denying the difficulties in articulation this sound brings about, I am just contending that students can self-monitor the production of this sound or an approximate target more consistently than the /ð, d/ distinction which, in my experience, remains a consistent "mistake" even after systematised.)

I am teaching advanced learners of English, with many years of exposure to the language -at least inside the classroom-, and with a certain level of proficiency in grammar, lexis and fluency. It comes as no surprise that these students have made their own hypotheses of what the language sounds like, and many of their "mistakes" are, in fact, developmental in nature. This is in-keeping with Major's (1987) Ontogeny Model that predicts the increase of developmental over transfer processes over time.

Adapted from Mayor (1987) in Ioup and Weinberger (1987)

So this is what many of my students produce:

In d̪e d̪ark! or In de dark!
d̪ey are d̪ancing und̪er d̪e stars! or dey are dancing under de stars!

Most students who master the plosive and alveolar aspect of English /d/ overapply it to cover the contexts for /ð/ as well. This is known as hypercorrection, and it counts as evidence that students are actually learning and making judgements as to how the language works! :D

"Tips and Tricks" to teach /ð, d/

The obvious, first: teaching the spellings of both sounds. But the presence of spellings alone is not enough to ensure learners' self-regulation in the production. So some extra tips may be presented to correct or give feedback in order to allow students to realise they should be producing a different sound:

Hand gestures + articulatory tips and tricks: I like using two gestures to represent these sounds (These gestures are not offensive in my culture, but they may be in yours or your learners', so please check beforehand! No offence intended here!)

This is to ask students to produce an interdental sound rather than dental, since in this way, my students manage to produce a better defined /ð/ sound, and they become aware of the "tickling" sensation the friction produces. (It may always backfire, since some students pull their tongues out too much and this obstaculises a smooth articulation, but all in all, the interdental option works pretty well). I have also found it that with all voiced fricatives I needed to help my students to breathe "properly", encouraging diaphragmatic breathing over their usual "chest-lifting" (needed a word for it!) breathing. Many of my students improved on their production of /ð/  and other voiced fricatives when adopting a straight posture, and pulling the upper part of their "bellies" (just below their ribs) outwards when they produced the sound, somehow increasing the amount of air employed for a smooth friction to take place in the mouth organs. This, in turn, allowed them to loosen some unwanted tension on the neck and jaws, as their attention was placed onto the actual breathing process.

This is to remind students of /d/. I have chosen to focus on either keeping the teeth "softly clenched" or the back molars for the production of /d/. In spite of the fact that the transition from Spanish dental to English alveolar merely implies placing the tip and blade of the tongue further back in the mouth (going from just behind the teeth to the "balcony of the mouth", the alveolar ridge), I have found that the more I ask my students to focus on the tongue for this sound in particular, the more the tongue "reacts" by getting into the dental position. So I have found that asking my students to build a "fortress" in the mouth with the teeth or the molars prevented them from "going dental". (This is not a tip I may exploit with all students, as many students have a very stiff jaw due to stress or tension, and this slight "teeth-clenching" may "spoil" their overall performance and create some distortion in sounds. I'll talk about "stiff jaws" and "flabby tongues" and their effect on articulation and voice quality in some other post)

Some other common tips for /d/ include (sorry if I could not trace the sources here):

• Exploration: exploring the teeth ridge area, playing with the tip of the tongue in different positions from the back of the teeth to the post alveolar area, and attempting different sounds till the right alveolar articulation is found. It allows students to become familiar with the alveolar area, a bit "underused" in Riverplate Spanish. It would sound something like this:

• Marking the territory: I remember reading somewhere about the suggestion that students could define the alveolar area by either putting a finger or pencil or straw behind their teeth (maybe not too hygienic!) to make sure students' tongues produce contact with the alveolar ridge only, or a piece of chewing gum on the right articulatory point (this is a contribution by my colleague Andrea P.) to identify the "right spot". Some people would say this is a bit "invasive", but if treated as a game and in the right conditions, it should be all right.

Now a few tips for /ð/:

• The "tongue" greeting": my trainer always presented this tip which is useful in kinder or primary classes. It involves travelling to an imaginary country, where people greet each other by showing the tip of their tongues, and students walk around the classroom saying "th" as a means of saying "hi" to each other.
• The "tickling" competition: students may be asked to pull the very tip of their tongue between their upper and lower tip and blow slowly, producing a "tickling" sensation. After a short while, this feeling begins to "bother" the lips, the friction and the vibration it produces at times makes students drop the sound. So the competition involves trying to "put up" with the tickling for as long as you can. This is just to draw students' awareness of friction, and what it involves.

There are some videos and tutorials on the web for this sound:

• Two home-made videos I made for my students: /d/ and /ð/ (used to looking ridiculous on the web by now! Sorry about the sound quality!)

Plus:

• BBC Learning Tips, /d/ and /ð/
• Some other videos for Spanish speakers (here and here). Some may be more "effective" and "accurate" than others. You will decide!

Hope you have found this post useful!

References

• Quilis, Antonio (1993). Tratado de Fonología y Fonética Españolas. Madrid: Gredos
• Mott, Brian. (2012) English Phonetics and Phonology for Spanish Speakers. UB: Spain.
• Moyano, G. (1996) A Comparison between English and Spanish consonants. Buenos Aires.
• Major, R.(1987) A Model For Interlanguage Phonology. In Ioup and Weinberger (1987) Interlanguage Phonology. The Acquistion of a Second Language Sound System.Cambridge: Newbury House Publishers
• Monroy Casas, R. (2001) Profiling The Phonological Processes Shaping The Fossilised Il Of Adult Spanish Learners Of English As Foreign Language. Some Theoretical Implications. University of Murcia
• Finch, D. and Ortiz Lira, H.(1982) A Course in English Phonetics for Spanish Speakers.London: Heinemann
• García Jurado, María Amalia y Arenas (2005) La fonética del español: análisis e investigación de los sonidos del habla. Buenos Aires: Quórum
• Ioup, G., & Weinberger, S. (1987). Interlanguage phonology: The acquisition of a second language sound system (Vol. 8). Newbury House.

Teaching Segments #1: English /ð, d/ - Part 1: Exploring phonemic and allophonic differences

I have been teaching general English and English "for international exams" (OMG, I do have a grudge for that) for over ten years now, and I started teaching phonetics at College in 2006. I wouldn't say I'm tooo experienced, but I would say that one of the features I have "corrected" and "systematised" the most among my Spanish-speaking students, is the contrast /ð/,/d/.

Some basic, general remarks for General British /ð/,/d/.:

• General British (GB) has a phonemic contrast between /ð/,/d/. This means that if we apply the process of commutation, through which we change one phoneme of a word for another, then we will have another acceptable English word, creating a minimal pair: "they"/"day", /ðeɪ//deɪ/. This proves that these two are phonemes of English.
• /ð/,/d/ are represented by different spellings in GB as well: /ð/ is generally, if not always, "th", /d/ is unproblematically spelled with "d" or "dd".
• /ð/,/d/ are clearly different in terms of place and manner of articulation, but share the same voice: both sounds are produced with vocal fold vibration. Curious to see what that looks like? Try this video. 
• /d/ is an alveolar sound in terms of its place of articulation, i.e., the tip and blade of the tongue are in contact with the alveolar ridge:

Image credit: http://www.study-languages-online.com/images/sounds/sound-d.gif

•  As far as manner of articulation is concerned, it is a plosive sound, which means that it is produced in three stages: a closure stage, whereby you put your articulators in firm contact, a compression stage in which the air is kept "hostage" behind the closure, and a sudden, "explosive" release (If English is your L1, or you have already "acquired" this sound, you can try this in slow motion). This is generally represented with parametric diagrams like the one below (adapted from Ortiz Lira & Finch, 1982):

• /ð/ is a dental sound, as the tongue articulates with the teeth (Mott 2012:134 states that in English it is generally "dental rather than interdental", though it is more useful in pedagogical terms to go for an interdental articulation, in my humble opinion!). This dental articulation is made fun of in the famous cartoon from the book How to be British:

Regarding manner of articulation, /ð/ is a fricative sound, and instead of presenting any degree of closure, there is a narrow opening between the articulators which creates friction (i.e, "turbulence"):

Fricatives. (Diagram adapted from Ortiz Lira & Finch, 1982)

You can watch an animation of their articulation in the University of Iowa site.

 (Digression alert!) Check out this MRI of a person singing a rap song (It's awesome!). It's great to see what goes on in your mouth cavity as you talk in real time!

What do we know about "d"-related sounds in my Riverplate variety of Spanish?

• We have got only one phoneme, /d/. As García Jurado and Arenas (2005:99) explain, it is an (post)dental plosive ("oclusiva apicodental")

Image credit: University of Iowa-Phonetics http://www.uiowa.edu/~acadtech/phonetics/spanish/frameset.html

Now. The interesting thing about our Spanish/d/ is that it has two main allophones, that is, two main variants. If you listen to my rendering of the word "dedo" ("finger"), you may notice two different productions of the sound (I'm still at odds with PRAAT's text grid feature...sorry!): 

1. This goes to say that my Riverplate Spanish /d/ has two "versions": a dental plosive [d̪]; and a dental fricative (even approximant at times for some speakers!) [ð], which appears in intervocalic position, and sometimes after nasals as well: ['d̪eðo].

(Not after nasals, my colleague Zabala corrects me. In my articulation it is true it is not a fricative, though I recall having read this somewhere. I'll trace the source)

So the RS learners of General British finds themselves entangled in what is called an allophonic split (Lado, 1957), that is, two allophones of their L1 which are two separate phonemes in their L2 (a "phonemization of allophonic variation", Moore, 2013).

Now. Is it only this split that makes /d, ð/ difficult for Argentinian learners of English? After all, how truly aware are we of the sounds we use in our language, as a general rule?

The answer to this question can to a certain extent, be found in a Contrastive Analysis (Weinreich, 1953), like the one carried out just now. But there may be other psycholinguistic factors also operating which may affect the mispronounciation of /d, ð/ among my learners . Part 2 will explore these phonological processes in more detail, explain some of the most common problems RS learners of English face regarding these sounds, and present some tips and tricks for classroom use.

Pronunciation and I: Intrinsic (!?) Motivation (a bit of a Freudian session)

Pronunciation has been my passion for many years now. I would feel tempted to say it was because of my Phonetics I trainer, Prof. Graciela Moyano, and I would say "certainly, yes!". And I would not be wrong. I am determined to devote a whole post to my wonderful training in Phonetics by Graciela. However, I know there is something about my curiosity about pronunciation that goes back to very early in my life, probably when I was 5 and I was sent to a speech therapist because I could not produce my "rr"s (BTW, I am a native speaker of Riverplate Spanish.  Curious what that means? Wikipedia may enlighten you...a little bit).

A bit of theory first. The "rr" sound is called a voiced alveolar trill, usually known as a "rolling r", and represented thus: [r]. It is a liquid consonant (though Cruttenden 2014:51 points out there is no common agreement on this) and it is a trill because it is produced with intermittent closure between the tip and blade of the tongue with the alveolar ridge. It is a very common sound in Spanish, represented by spelling "r" (initial in the word) and "rr". You can see an animation for the articulation of this sound in the University of Iowa website: http://www.uiowa.edu/~acadtech/phonetics/spanish/frameset.html (Look it up under "vibrantes"). 
You can hear the English frictionless continuant approximant and the Spanish alveolar trill (together with other manifestations of /r/) here. 
This is what the spectogram of my Spanish [r] and my English [ɹ] look like in PRAAT:

And what they sound like: 
(BTW, for some reason at times (not here, I think) my English /r/ includes a labiodental articulation, very much like Johnathan Ross' own [ʋ] and I guess I was trying to avoid it here. I wonder why!)

Here's a video that shows some people attempting to produce a Spanish trill (BTW, that was my "rr" at the age of 5! My Italian granny used to call me "my little Frenchy", at least that was what my /r/ sounded like to her, much to her dismay) and some tips and tricks which may (?) work (I don't really think the protagonist's achieved "rr" is truly alveolar, as there is some sort of velar start to my ear...but I may be wrong!):

You will hear it in Scottish, as this video shows:

I remember my speech therapy sessions quite vividly, which I guess is quite telling. I can recall disliking the speech therapist, a very tall, slender woman, and I get the feeling she was not very patient. I remember she filled a notebook with drawings and words with "rr", nursery rhymes as well: "Erre con erre cigarro, erre con erre barril..." and I was expected to repeat. I don't have any recollection of having been asked to do anything else than repeat. 

After six or seven sessions, she told my mother I had to keep attending sessions, even during the summer break. I remember getting in a tantrum, saying I was not going to go on attending, and my mother politely saying that since I was going to start swimming lessons, I would not be going back to my speech therapy sessions again. I was relieved.

I eventually got the sound "rr" on my own. I was probably not ready for my /r/ before this time (after all, it is one of the last sounds to be acquired,) or there had been, very likely, more articulatory practice in the sessions which my subconscious mind has "blocked".

***

I heard someone say that we choose our profession as a means of healing a past trauma. I don't know if this is the case (this experience competes with the fact that I was locked inside an English Institute when I was 8), but I do believe this was a significant event in my life and there is something about the impression that I was only asked to "repeat" which I react against, now, in my teaching. (Worth another post!)

Thank you for reading me! Any personal experiences to share regarding your /r/ sounds?