“Chimeras” of Spanish phonetics

I recently came across a fascinating research paper by Eugenio Martínez-Celdrán of the Universitat de Barcelona about “chimeras” of Spanish phonetics. A chimera is, essentially, an unrealistic notion. The paper presents three case studies in which traditional phonetic description of the Spanish sound system has under- or overshot the mark.

The three case studies include:

  1. Stops, fricatives, and approximants (a discussion of the voiced series /b, d, g/)
  2. Devoicing and voicing (primarily phonetic outcomes of /p, t, k/)
  3. Apico-dental vs laminal-denti-alveolar (phoneme /d/)

The paper’s main strength is its empirical foundation: arguments are well supported by sound spectra and palatography. A secondary strength is its implicit reminder that the phonetics of any language is much less tidy than the teacher or student might wish it to be, and viewpoints are often held out of convenience rather than because they reflect empirical conclusions. Hence the need for edited volumes such as the one from which this article is excerpted:  Selected Proceedings of the 3rd Conference on Laboratory Approaches to Spanish Phonology (2008), edited by Laura Colantoni and Jeffrey Steele (Cascadilla). Over the past decade or so, the field of “laboratory phonology” has grown monumentally, due to an upswing in interest in accurate phonological description as well as increasingly easy access to “laboratory” equipment by ordinary individuals rather than well-funded university departments. For example, sound spectrography, a longstanding keystone of laboratory phonological analysis, can now be done on just about any laptop using free downloadable software.

Of the three chimeras Martínez-Celdrán covers, I find the first one most applicable to a pronunciation lesson. The Spanish voiced obstruent series /b, d, g/ includes three allophones, traditionally fricatives [β, δ, γ], which pervade spoken Spanish and are therefore essential to a good “accent.” Martínez-Celdrán rounds up evidence that the three allophones are better described as approximants— i.e. more like [j] and [w] than like [v] and [z]. Over the years, various phonetics classes of mine have applied playful nicknames to these allophones, such as “leaky” and “relaxed.” On sillier days it has even been proposed that accurate pronunciation of the series requires a decent level of intoxication to ensure that the vocal apparatus (particularly the tongue) is adequately relaxed. Indeed, the designation “fricative” does not seem right, since even the less observant student knows that the pronunciation of nada as [náδa] is not commonly heard. In light of this case study, the descriptors “leaky” and “relaxed” are far more accurate than we thought and, in fact, may be the most practical way to describe an approximant without actually delving into thick articulatory terminology, formant structure, palatography, etc.

A good review day

At the end of a good review day, the classroom whiteboard might look like this: a veritable work of art of phonemic and phonetic transcription. Of course, it takes a lot of dry-erase markers to support this particular habit. Markers for all the students!  But the payoff is high. By putting transcription assignments on the board, students re-perform key transcription tasks and are then able to view the work of others, ultimately connecting the visual (transcription) with the auditory (pronunciation).

FullSizeRender (1)

IMG_0814  IMG_0815

Do phonetics classes lend themselves to standards-based grading?

To be sure, I was quite skeptical the first time I read the phrase standards-based grading. My first thought was that it was just a fancy way to water down standards without anyone caring as much– a positive “spin” on dulling the curriculum to appease mediocre students or to accommodate mediocre teaching. On a less cynical level, I was also puzzled. If “traditional” grading isn’t standards-based, then what is it?

After doing some homework on standards-based grading (hereinafter SBG), I found that it is actually quite well suited to teaching Spanish phonetics or, I would imagine, the phonetics of any language. Below I explain why. But first I’ll explain what exactly SBG is.

Most importantly, SBG does not actually dispense with grades. Rather, it aims to give a truer measure of achievement than traditional grading (hereinafter TG). In TG, many grades earned over time are averaged to come up with One Big Grade that represents student performance. However, there is plenty of evidence (much of it simply intuitive) that measuring student performance–or anyone’s performance for that matter–does not necessarily provide the best measure of learning achievement over a given time period.

In SBG as I understand it, a student essentially receives no binding grade until the last one, which measures their benchmark of achievement. All previous grades are merely advisory; they do not (need not) figure into the student’s grade. WHAT? I hear you ask (so did I). Please read on…

In most life experiences involving trajectories of learning and improvement over time, the only measure that counts is the most recent (or final) one. For example, in a performance evaluation, both employee and employer would concur that the employee’s most recent evaluation is the best measure of the employee’s achievement. The employee would find it quite unfair if their evaluation was “averaged” over the past ten years. Likewise for airplane pilots and lawyers: the only exam that counts is the last/most recent/latest/final one, because that is the one that shows whether or not learning requirements were met. Previous attempts are not “averaged in.” Given all of these intuitively satisfying precedents for SBG, why is it not widely implemented in the classroom?

My answer: Perhaps it should be.

It appears that successful implementation of SBG in the classroom requires creative thinking and full deployment of one’s teaching skill– both of which are mentally and intellectually taxing. Thus it will come as good news that SBG doesn’t seem to require any major changes to a course’s structural integrity. For example, twenty quizzes and five unit exams and a final exam can still be given and graded. The only difference is that all grades before the final grade are advisory in nature: demonstrating to the student areas of weakness and strength.

On the other hand, SBG does seem to require a rethinking of the course’s content and format. For example, the best way for the final measure to be a true one is for the course content to be cumulative, i.e. each new topic must build upon topics already learned. Each quiz or exam, ideally, must be at least partially cumulative, recycling old material. Otherwise, the final measure would only assess a portion of the course content covered. In a course in which material is sequenced compartmentally, this would be problem.

In a phonetics course such as the one I teach, material is explored and learning is assessed cumulatively. I consider this partly the nature of the subject matter, partly the way I assess learning of it. There is no easy way, for example, to explore consonant allophony without simultaneously visiting (or revisiting) vowel allophony, because the two always coincide in real speech, and therefore co-occur freely in transcription. Thus my Spanish phonetics class is already formatted for SBG; I just need to transition my thinking and grading methods, and transform the classroom culture (no doubt all of these are easier said than done!). My written exams involve strictly textual transcription (students transcribe a written text phonemically and phonetically), so each of the three “unit” exams builds upon previously learned material. The retooling: make exams 1 and 2 “advisory” and make exam 3 the written exam that “counts.”

Of course, there are quite a few arguments that an experienced teacher will readily levy against SBG; for example, what about students who drag their feet during the term because all those waypoint grades “don’t count?” Or what about students who “don’t test well?” Isn’t it cruel to pin the course grade (ostensibly) on a single assessment? My quick counter-counter-argument is that these matters don’t have anything to do with how the student is graded; rather, they are academic management issues that a teacher needs to be addressing effectively already. If a teacher isn’t already addressing the conjoined problem of weak exam takers and noncommitted students in their classes, then they would need to begin doing so. But this is hardly an SBG problem.

For more information on SBG, I highly recommend Alfie Kohn’s informative blog on the subject, as well as this very good overview by Iowa educator Mike Townsley.

What are your thoughts?

“Book people” a la Fahrenheit 451, and the importance of memorizing for articulatory accuracy

The big question with teaching pronunciation is how to grade students on their performance. There is material for several blog entries there, but I will touch one just one point, which is actually a question: “Read or memorize?”

In years past, I gave students a text to read aloud as their final oral presentation. They did the final reading by appointment in my office. I soon realized, however, that no matter how deeply they had mastered all of the phrase-level intricacies, as soon as they were reading from the printed page, they reverted to talking like SIRI.

Aha! For smooth, fluent speech, reading a printed text is a distraction and a hindrance! Lesson learned.

I now require students to memorize a text. The text I use is roughly the first two paragraphs of Laura Freixas’s Final absurdo. The text is highly visual and cinematic, has a smooth yet varied flow, and is fairly easy to memorize. It consists of roughly 16 distinct phonic phrases, and pretty much all the phonemes of Spanish are represented. Most importantly of all, the text really lends itself to smooth-speech finesse.

Students receive their copy of Final absurdo on day one of the course, and we “recite” it in unison as a class warm-up every day for the duration of the semester. After the third week, reading the text during the recitation is no longer permitted (i.e., they either have to memorize it or just “mumble along”). To ensure that there are no stragglers, around the third week I ask each student to come to my office and do a “content recitation” (not a “technical recitation”) for a grade. All I grade them on is memorization. The technical recitation-“The Big One”-stands in for a written final exam and occurs at the very end of the semester.

To enliven the perhaps daunting task of memorizing a Spanish text, I introduce my students to Esperanza García’s eloquently delivered recitation from Gabriel García Márquez’s Cien años de soledad. Esperanza García is a “persona libro” in true Fahrenheit 451 fashion: “Soy el primer capítulo de Cien años de soledad,” she tells aviondepapel.tv, a Spanish literary blog. Her memorized recitation of the novel’s famous opener (see video clip below) is instructive in its own right, as she makes extensive use of rhythm and intonation to emphasize the story’s information structure, yet (of course) does so fluently and naturally.

All of this to say that my students’ from-memory recitation of Final absurdo yields a far more accurate picture of their achievement than text read off the printed page, and allows for far greater expressiveness.

The real reason(s) for music

From as far back as I can remember, pop songs have been a cornerstone of introductory Spanish classes. Originally requiring a tape recorder or playback device, almost any Spanish-language song can now be accessed instantaneously on YouTube and played over the sound system of a modern classroom with ease.

Of course, the reasons for using songs in class weren’t always clear. Probably because the pedagogical benefits of songs in class weren’t always clear. Sure, there is this vague sense that a second language processed as music does something to the right brain, or works synergistically in some way with one’s artistic skill, or some such. Honestly, I think the main reason songs get used so much is because they’re fun and give students a break from the humdrum of grammar drills.

Fortunately the science behind the effect of music on language ability is becoming continually clearer. Since my niche is teaching Spanish pronunciation to native English speakers, I find the connections between music and pronunciation especially interesting. Most studies look at the positive effect of music–for example, singing–on various fluency impairments, such as stuttering, autism, and aphasia. These studies focus primarily on the positive effect gained in one’s first language. A (2010) article in the journal  Musical Perception titled “The therapeutic effect of singing on neurological disorders” explores the neurological benefits of singing (with extensive literature review), but acknowledges that the reasons for the benefits are not yet clearly understood.

The fact that autistic children who are unable to speak are often able to sing with relative ease suggests that singing mediates fluency. Teachers of pronunciation, take note.

The musical benefit on language holds true also for second language pronunciation where no neurological impairment is present. One study found that Taiwanese children who underwent a regimen of “music instruction” (singing songs, rhymes, chants) as a part of an English course had statistically improved pronunciation compared to children in a control group. The music instruction also benefitted vocabulary retention.

Perhaps to my shame, the singing component of my Spanish pronunciation courses is a relatively new fixture. But it has become a permanent and foundational one. Students send me the YouTube links to Spanish-language songs (I ask that the songs be sung by native Spanish speakers born outside the U.S.), I prepare a lyric sheet, and we listen to the songs and sing along with them for the first 10-15 minutes of each class. Songs are rotated frequently so there is always something new in front of the students. The results bear fruit, though not immediately. The style of music does not appear to matter, though songs with several repeating estribillos are more likely to get students actually singing, rather than just mumbling or reciting the lyrics silently in their heads.

Nothing like songs in Spanish class every day of the year, without the guilt that we’re just doing it to kill time or–heaven forbid–to have fun.

SIRI and the challenge of fluency

Most second language learners of Spanish (or any language) go through a phase in which their spoken language is based upon strings of printed words, complete with all the spaces that come between them. Thus Los otros están en el avión might be pronounced [los…ó–tros…es–tán…en…el…a–byón]. This makes sense, because that is how teachers coddlingly pronounce the language: it’s easier to understand a spoken phrase with plenty of comfy “daylight” between the words! Not surprisingly then, a fluent speech pronunciation [lo–só–tro–ses–tá–ne–ne–la–byón] comes as a bit of a shock. Even more so its phonetic representation, with all the words “running together” in accordance with phrase-level syllabification rules.

The phonetic principle involved strikes many students as a foible of phonetic transcription rather than a fact of pronunciation, with more than a few students noting that they “would never say” Los otros están en el avión in that way. Indeed, much of teaching phrase-level fluency lies in convincing students that there really is such a thing as phrase-level phonetics, and that they hear phrase-level phonetics in action whenever they hear naturally spoken Spanish.

Equally important is the generalization that the word-by-word pronunciation isn’t really wrong, just perhaps a little too precise for natural conversation.

In a recent class of mine, a few students weren’t convinced that the difference between word-by-word pronunciation and phrase pronunciation was significant enough to worry about. To many students, the two “sounded the same.” So what was the big deal? Then came one student’s ingenious comparison to SIRI.

SIRI, of course, is the Apple iPhone “voice” that answers basic questions and occasionally turns sassy. Kudos to my student for pointing out that SIRI constructs her sentences word-by-word, and that is the main reason (perhaps the only reason) that her speech sounds odd to the ears of an English speaker. Not wrong, just… odd. Nods of agreement all around, as understanding dawned. The class quickly agreed that SIRI sounds “robotic”, “stilted”, “over-precise”, “forced”, etc., and that if her word-by-word speech pattern isn’t natural-sounding English, then word-by-word pronunciation of Spanish must not be natural-sounding either.

Point made.

Telemundo, Spanish pronunciation, and TV ratings

One of the basic questions the pronunciation teacher must wrangle with before the first day of class is which variety of Spanish to teach. While it might seem like a forward-thinking idea to teach students “everything they know” about Spanish dialectology (i.e., everything from the nuances of Castilian Spanish to such rich variants as habanero or the porteño of Buenos Aires), this Washington Post article (originally published August 2, 2004) makes a strong case for emphasizing Latin American standard Spanish to the exclusion of other varieties. The article looks primarily at how Telemundo, a major Spanish-language cable channel, promotes the usage of “Mexican” Spanish as its corporate standard because this variety results in higher viewer ratings. Of course, “Mexican” is just shorthand for norma culta, or Latin American standard pronunciation, which captures features not only of Mexican Spanish, but also of limeño Peruvian, bogotano Colombian, and others. The historical basis for the norma culta is summarized in Appendix B of my book, and the phonetic details can be reviewed here and investigated quite thoroughly in John M. Lipski’s important book Latin American Spanish (Longman, 1994).

Students are typically intrigued by the social implications of Telemundo‘s empirical ratings. Furthermore, they grasp the reasoning that while there may be as many varieties of Spanish as one might be able to count, not all of them are equal from a social or even a practical perspective. Indeed, the mastery of just one–the Latin American norma culta–opens a world of doors in Latin America.