Printer Friendly

Linguistic lingo and lyric diction III--pronunciation contrasts.

The reader who has engaged in studio voice pedagogy probably on occasion will have taught a student whose first language is not English, and who has persistent difficulty in pinpointing the precise formulation of an English sound, either in speech, in singing, or more likely in both. This is a natural occurrence for any EsL student, because every human dialect has its own unique set of phonological principles and traditions that form a received set of standard pronunciation reflexes among the speakers of that dialect. Everyone inevitably hears another language in terms of the phonemes of one's native tongue--or, to mix metaphors, auditory tinted glasses. The pedagogic challenge of teaching ESL learners idiomatic pronunciation and lyric diction can be persistent over a long period, and can even at times appear insoluble, so ingrained are the muscular reflexes to adopt a familiar articulation that is incorrect in English. Such differences are typically obvious to the English-speaking instructor, and can be elusive to the point of exasperation for the ESL student. "I just don't hear the difference" is a common reaction for any student, and extends even to those whose native tongue is English. Several factors are involved.

The propensity to hear subtle differences in articulations can vary from one individual to another on a purely phonetic level, independent of language. This can only partly be explained as a product of experience and training. Linguists distinguish between three levels of sound perception: acoustic, articulative, and perceptual/cognitive. When working on lyric diction with a student, the instructor naturally does not pause to consider on which of these three levels the conversation may be proceeding; nevertheless, one of these levels usually implicitly predominates. The differences between them are not recondite or theoretical. Attempts to improve pronunciation by zeroing in on tongue placement, lip rounding, openness, etc. are articulatively based, as are most discussions of aggiustamento strategies. Two identically uttered articulations by the same person can result in different spectrum images on an analyzer when the room acoustic is altered. of course, the instructor usually has no control over this factor in a typical lesson or coaching. Every teacher is familiar with the ways in which a singer's technical approach to a song or aria can be drastically altered by moving from a dry small studio into a resonant church acoustic, or from an acoustically well balanced studio into a cavernous concert hall in which performers sense nothing "coming back at them." The strategies for overcoming these challenges will center upon technical strategies of voice production ("don't push," "trust that the sound isn't dropping to the floor six inches in front of you," and the like). But acoustic changes can do odd things to phonetic clarity as well. The most overt, ubiquitous challenge here is enunciation, and the frequent need to exaggerate in comparison with the teaching studio. But vowels and consonants can be altered acoustically and perceptually too in different acoustics, because of the alterations in the relative prominence of overtones that result from each space. Singers perceive their own voices through a combination of external and internal stimuli, and will sound different to themselves on recording for this reason, as well as from the change to electronic transmission.

One effective way to for singers and pedagogues to learn about the sound pattern differences between and two or more languages is to perform a systematic contrastive analysis of them. In the 1960s a series of contrastive structure monographs was published by the University of Chicago Press, illustrating the phonological behavior of German, Italian, and Spanish by comparing their sound systems with that of American English. As stated by the authors in the preface to the Spanish volume, the justification for the book is that "people have trouble learning other languages because they speak their own so well." (1) There was nothing new in the idea of learning about the phonology of a language by systematically comparing it with another, presumably familiar one. As early as 1730 in London, Claudius Arnoux had published Parallels of the Sounds and Syllables of the French and English Languages. Some years after the appearance of the University of Chicago series, Don and Alleen Nilsen published a thin, unassuming practical manual, Pronunciation Contrasts in English, as an aid to teachers of ESL. (2) Since that time, the market has seen several comparable manuals designed to streamline the acquisition of English through the use of contrasting minimal pairs. Unlike the earlier series, this manual illustrated the phonemes of English by contrasting sounds within English itself, limiting itself to phoneme pairs deemed to be phonetically similar enough to engender confusion. The page contrasting [e] <bet> and [ae] <bat> will serve to illustrate. (3)

Sentences with contextual clues-
   This BED is BAD.
   I GUESS they want GAS.
   The GEM fell in the JAM.
   What she SAID made me SAD.
   Can you BEND this iron BAND?
   Minimal sentences-
   SEND / SAND it carefully.
   The MEN / MAN will come.
   The PEN / PAN leaks.
   He got hurt on the TREK / TRACK.

She didn't want to talk about the PEST / PAST. Such contextualized pairs, of course, serve to reflect the importance of distinguishing between such vowel phonemes, because of the resultant differences in meaning. A lengthy list of further minimal pair words for each pair of phonemes is included on each page. Thirty-two English vowel pairs and forty-two consonant and glide pairs are contrasted in the book. Perhaps the most valuable data on each page for the singing teacher, though, is the list of languages whose speakers can be expected to have difficulty, in varying degrees, with each particular pair. More than fifty language specialists were consulted in the formation of these lists. In our example above ([e] vs. [ae]), for instance, forty languages whose native speakers would be expected to have difficulty with this contrast are listed.

The consonants [??] <thin> and [??] <this> provide a good illustration of the approach for consonants, as they exist in few languages other than English. There are pages contrasting [??] thank with [t] tank, [??] shank and [s] sank, while [??] thy (than, then) is contrasted with [??] thigh, [d] Dan, and [z] Zen. As expected, the marked nature of English [??] and [??] results in some of the longest language lists in the book. German appears on most of these lists, since [??] and [??] are absent in German. This tallies with the substitutions [t], [s], [d], and [z] commonly encountered in German accents in English.

It was unfortunately beyond the scope of the Nilsen book to itemize specifically why each language would find the contrast troublesome. Such information would be of great benefit to both ESL and singing teachers. There are two primary reasons why a sound can prove difficult to acquire and/or differentiate from other sounds. The more obvious reason is that the sound may be nonexistent in the home language. But an equally common and more troublesome reason is that a sound may exist in the home language, but be employed in different environments. This can mean not only that the sound will tend to be avoided in places where it should occur, but also that it may show up in incorrect environments instead, as paralleled in the home language. (4)

The old Lakota proverb, "Never criticize someone else until you walk a mile in his moccasins," outmoded though it may now be in its verbiage, is singularly appropriate advice for the voice pedagogue. The best way to anticipate and address pronunciation challenges for ESL students is to understand from the outset the phonological system of the student's native language, and to take note of the differences between it and English in terms of its inventory of articulations and the contrasting environments in which each is used.

In the singing studio, contrasts between languages most often employed in the standard languages of the Western musical canon will be generally, if not systematically, appreciated by the instructor. But the reliance on familiarity and instinct quickly breaks down once one moves beyond the standard five or six. Understanding the contrasts between English and such European languages as Dutch, Swedish, Norwegian, and the various Slavic languages of eastern Europe will be the exception rather than the rule among voice instructors. The same can be said of languages beyond the Indo-European family, where the differences in all aspects of the grammar (including the sound system) are usually much more substantial.

When an instructor works with a student whose first language is unfamiliar to her or him, the present author thus recommends that a basic familiarity with the sound pattern of that language be acquired by the instructor as a prerequisite to effective pedagogy. This sounds far more daunting than it actually is. The aim is not so much to attain comprehension of grammar, vocabulary, and fluency as it is to become familiar with the sounds and basic behavior of the limited number of articulations employed in the language. Charts, tables, and brief prose guides to the phonology of virtually any language can now be obtained online, and in standard reference works. (5) A three-step process will aid in acquiring basic phonological skills quickly.

1. With the aid of vowel and consonant diagrams and charts, a quick comparison can be made between any two or more languages, to determine which articulations are common to both, and which are specific to one and not to others.

2. An inventory of phonemes for each language, with all possible spellings for each sound, will illustrate how the sound patterns are realized in writing. Such an inventory may begin with the phonemes, and list all orthographic renderings of that sound. A thorough survey will also describe the distribution of allophones for each phoneme. Conversely, an inventory may list all occurring letters and clusters alphabetically, providing all possible articulations encountered for each, and the environments in which they occur. Because singers usually encounter repertory first through the sheet music with its printed text, it is important to understand the particular relationship each language has between its sound pattern and its orthographic conventions. Unfortunately, such lists are often lengthy, intricate, and with limited overlap of vowel and consonant clusters between languages, making systematic contrastive analysis of this aspect of language unwieldy. Moreover, useful lists of this sort are often hard to find or quite unavailable in reference sources in English, especially for less familiar languages.

3. Further reading on the phonological characteristics of the language concerned, to be aware of other contrasts not covered by the first two steps, is also essential. Particularly important at this stage is an awareness of the behavior of the allophonic variants of phonemes, if not covered above, and how the same variant may distribute itself quite differently between any two languages. An articulation that is phonemic in one language could well be allophonic in another. This is one of the most prevalent sources of confusion and difficulty in acquiring mastery of any unfamiliar language.

The ordering of these three steps is quite deliberate, and they have been ranked in procedural order, which tends to correspond also to the ready availability of information in published and online sources. As with step 2, finding accurate and thorough descriptions in English of the sound pattern of a less commonly encountered language can be elusive.

To see how this process might operate in real terms, we shall choose two languages for comparison with American English-one encountered in standard vocal repertory (French), and one that is likely to be less familiar for most readers or quite unfamiliar, at least in singing (Dutch). Observations derived from all three of the steps described above will be included. It will thus be seen how a limited but relevant body of information about a language can be of considerable practical assistance to the voice pedagogue who lacks background in the language concerned. French will exemplify a language quite familiar in the voice studio, to illustrate the contrastive approach. Dutch will illustrate how that approach can be extended to any other language. Abbreviations D (Dutch), E (English), F (French), G (German), and I (Italian) will be employed. The author intends to extend this demonstration to languages outside the Indo-European group in a future article, to demonstrate the universal applicability of this process.


A--Segmental (step 1)

A comparison of Tables 1 and 2 (Appendix) will reinforce the six most familiar phonological differences between French and English consonants.

1. F possesses a palatal nasal [n], which is often substituted for the velar [n] so common in E, but absent in F. As a result, these new articulations must be drilled, and reflex habits avoided. The words signe and sing illustrate the difference, and the distinction between E sing and sign can pose problems.

2. The French speaker has as much difficulty with English spoken [??] as E does with F [??]. This difficulty is largely neutralized in singing (except in popular genres), because flap [??] or trill [??] is standard in both languages.

3. The "dental" consonants [d], [n], [t], and [l] must be so in F, and E speakers must work toward making a more forward point of articulation routine.

4. The E dental fricatives [??] and [??] are well known stumbling blocks, as those articulations are not present in many European languages.

5. The [y] glide of F is elusive for many E speakers, and the closest familiar sound [w] is usually substituted until the new articulation becomes routine. Suite will sound like sweet until the [y] is mastered.

6. The common E glottal [?] is a marginal articulation in F, and will be avoided in singing, except in the most vehement, dramatic moments.

Similarly, a comparison of Figures 1 and 3 will reveal the many points of departure between E and F vowels. These differences involve both inventory and the placement of vowels common to both languages. Vowels that exist in only one of the two languages include E [ae], [i], [a], [??], and F [y], [??], [oe], [??] plus the four nasal vowels. This means that the F speaker must acquire four new vowel sounds in E, and the American E speaker must acquire eight new vowels in F. The closed [e] and [o] so common to F are marginal sounds in English, occurring only occasionally in unstressed syllables. Similarly, open [??] will be present in some dialects, and exist as a separate phoneme in British English, but is not encountered in standard American. The front [a] is standard in F, but found only in some dialects of English. Some E speakers will partially nasalize vowels, either as part of their regular speech habits, or alternate a nasal allophone in some environments with an oral one in others. But the four nasal vowels of F exist as phonemes distinct from their oral counterparts (pas/paon, dos/dont).

It will also be noticed that vowels with the same symbol are not necessarily articulated in exactly the same location between F and E, and rarely occupy the extreme cardinal vowel position. This is a more complex matter, harder to generalize upon, because of the inherent latitude in vowel color even within each language, from individual to individual or from one dialect to another. The schwa in particular is unstable, and has a pronounced tendency in F to veer toward [oe], especially in more prominent syllables. Notice in particular the brighter placement of [e] in F, and the more centralized [u] of E.

The tables exhibit for the most part only the phonemes of each language, and do not outline the many allophonic variants present in any language. In the case of consonants, this is because such tables isolate only two features of articulations: the place and the manner of articulation. Readers familiar with the basics of generative phonology will appreciate that many other features are present in sounds, beyond the grasp of two-dimensional charts. Such features are often the necessary tools in distinguishing allophonic variation, and are often dealt with in nomenclature with the use of diacritics, such as [ph] and [p[??]]. For a pedagogue interested in acquiring a working knowledge of the articulative behavior of an unfamiliar language, this represents a next stage in inquiry (steps 2 and 3). This will usually involve prose reading, rather than the convenient comparing of diagrams. While basic phonological depictions of virtually any language are readily available online and in books on languages of the world, digging down to this next level of comprehension for specific information is often more elusive.

B--Suprasegmental (steps 2 and 3)

Although a contrastive analysis of speech sounds in isolation between two languages as presented above is a good starting point, the manner in which sounds behave in connection with one another is essential to the acquisition of fully idiomatic pronunciation. Phrase and sentence phonology as well as phonotactics all play an important role. The native tongue of an ESL speaker is often discernable in one who mastered the sounds of English at only the segmental level. In singing, these telltale signs are often conveniently masked or fully neutralized. The natural rhythmic flow of any language (long/short vowels, strong/weak forms of clitics, and so on), for instance, is disguised by the rhythmic imperatives of the music itself. It is for this reason that singers may sound "authentic" in the sung moments of a stage work, and much less so in accompanying spoken dialogue, as in the German dialogue of Die Zauberflote, the spoken moments in La traviata, and in the operettas of Offenbach and Johann Strauss Jr. One cannot disguise the suprasegmental subtleties of another language in speech that way one can in singing.

Suprasegmental considerations in French that have implications for lyric diction include liaison practice, vowel harmonization, pronunciation of e-muet in style soutenu, elision, syllable and word stress, and absence of vowel reduction. Elaboration of these considerations can be found in several readily available sources, and is unnecessary to undertake for the purposes of this article. (6)


Dialect variation in Dutch is considerable, even across relatively small distances, such as the Dutch spoken in Belgium. Standard Netherlands Dutch (known as Hollands, after the province) will be the reference for our purposes here. It is sometimes referred to as ABN (Algemeen Beschaafd Nederlands). Because of the strong presence of English in the Low Countries, the majority of Dutch speakers will have at least some familiarity with the sound pattern of English, and the very different relationship between orthography and pronunciation that exists in each language. It is more likely that the voice instructor will be unfamiliar with the sound/spelling patterns of Dutch, and how different they are from English and German.

In keeping with step 1, one can contrast Table 1 with 3, Figure 1 with 4, and Figure 2 with 5 (see Appendix). From these one can make a number of deductions.

Tables 1 and 3

Any articulation present in 1 but not in 3 will have to be acquired by a D speaker. This includes the dental fricatives [??] and [??], the voiced velar plosive [g], the velar approximant [w], and the unvoiced glottal fricative [h]. One might find the absence of [g] in D surprising-and indeed further reading will be in order, to clarify the situation. Conversely, in D there is a labiodental approximant [??], the velar fricatives [x] and [Y], a voiced glottal fricative [h], and in parentheses a postalveolar plosive [c], a palatal nasal [n], and a voiced uvular trill [R]. (7) We notice also that four of the E alveolar consonants are dental in D, and that the postalveolar fricatives [??] and [3] are in parentheses in D. We are unsure at this point what the parentheses might signify.

Figures 1 and 4

The situation with D vowels is somewhat complex, and is different not only from English and French, but also from Hochdeutsch, with which Dutch has otherwise much in common. The vowel quadrilaterals evince considerable overlap of vowel sounds between E and D, although many are located in different places, as can be seen on the tables. [i], [1], [e], [a], [u], and [??] are common to both languages, and [e] and [o] could be added to the list, with conditions. D [e:] and [o:] are found on Figure 5, because of their tendency to diphthongize, and are uncommon in E because they are restricted to unstressed syllables. D has a full set of rounded front vowels, [y], [y], [??], and [oe], although they are more centralized than in F or G. (8) D speakers will need to take care not to introduce these mixed vowels into E. E [ae], [??], and [??] are not found in D, and will have to be acquired. In D there is an opposition between long and short [i], [y], [e], [oe], [??], and [u], while vowel length is not phonemic in E. The mid-low back [??] is shared with British English (law), but is found only in some American dialects. The schwa in D is not rhoticized-more like British than American E.

Figures 2 and 5

E has five closing diphthongs, three ending in [i] and two in [??]. The three wide diphthongs of D all close further than in E, closer to [i], [y], and [u]. The D diphthong that will sound least like E is [oey]. It is closest to [ai] and [[??]i], but still quite unlike either. This is the main source of difficulty in diphthongs, both from D to E and E to D. The E [a[??]] is more back in D. British E also has four centering diphthongs (not shown on the chart), ending in [??]. Although the schwa is a frequently occurring vowel in D, it does not form a component of any diphthong. A marked feature of D is the presence of three long mid-high narrow diphthongs (i.e., they travel less distance)--[e:(i)],[[??]:(y)], and [o:(u)]. More will be said about them below.

These tables are of course incomplete in the information they convey. One must move now to step 2 to know how the written language symbolizes the sound pattern. The following outline exhibits in a simplified form the type of information that can be acquired at this stage. One of the more vexing complexities of Dutch is the orthography, which frequently implies something quite different from its actual pronunciation, if E, F, G, and I are the frames of reference--as they usually are. For instance, D <ui> looks as if it might be [yi] as in F puis, or [u:i]/[u 'i:] as in Italian lui/Luigi. A Dutch speaker is likely to have trouble with E words like quick as a result. In Dutch orthography, <ui> is [cey], like lui, while <oei> comes closer to the I <ui>. A good way to practice the [oey] diphthong initially is to start with the more familiar D [ei], and lip-round. It is then just a matter of thinking of a slightly more centralized tongue position for [oey]. How Dutch vowel phonemes relate to orthography is shown in Example 1.

Double letter vowels are always long ("free"), while single letter vowels may be short or long. Generally, a single vowel will be long in an open syllable and word-final, and short in a closed syllable. Thus manen ['ma : nan], but mannen ['ma n[??]n]. Contrary to what one might expect, the short <a> is a back vowel [a], and the long vowel is front of center [a :]. Vowel length is phonemic in D, with several minimal pairs: boos/bos, vies/vis, veel/vel, and maan/man.

The long closed vowels [e:], [[??]:], and [o :] are known as "narrow" closing diphthongs, and are usually symbolized as pure vowels. The degree of sound change can vary, and some speakers simply employ pure long vowels. When diphthongized, it is never to the extent of the corresponding E diphthongs [ei] and [o[??]]. A Dutch singer is likely to substitute the narrow D diphthong for these, in words such as day and though. (9)

The Dutch "wide" diphthongs are distinct from those of both German and English, with the closest equivalents being
D         E             G

[[??]i]   [[??]i]       [a[??]] or [ae]
[oey]     --([a[??]])   --([[??]i] or [[??]Y])
[[??]u]   [a[??]]       [a[??]] or [ao]

The [ei] of D is a more "active" diphthong than the [ei] of E, in that it closes more, and treats each sound distinctly, much like in Italian. A speaker of D will have trouble with the E diphthongs in mate [ei] and might [ai], substituting the [ei] of Dutch mijt. The initial vowel of D [ei] is also distinct from that of E, as can be seen from their relative positions in the vowel charts. Dutch bij is arguably more like Italian bei than English bay or buy. Of the three diphthongs, [au] comes closest to E, as in D nou / E now.

Dutch also has two-syllable vowel sequences that are not familiar from English or German, except perhaps as diphthongs.
a:i    aai    maai
o:i    ooi    mooi
e:u    eeuw   meeuw
u.i    oei    roei
i.u    ieuw   nieuw
y:u    uw     ruw

Like E, D employs the schwa as a default articulation that vowels reduce to in unstressed syllables. Thus, the tendency in F and I to retain vowel quality regardless of syllable stress is not present in D, and should not pose a problem when moving from D to E or vice versa.

Dutch consonants are somewhat less complex than vowels, but involve a few articulations unfamiliar from E or G. These include the labio-dental approximant [??], the voiced velar fricative [Y], and the voiced glottal fricative [??] as found on Table 3. All can be heard initially, as in wang [[??] an], gaan [ya:n], and hoed [[??]ut]. The voiced/unvoiced symmetry of the plosives breaks down with the velars, since [g] is only to be encountered in loan words. Historically it changed to [x], and is pronounced thus in some loan words: F galant [xalant]. The voiced/unvoiced

distinction of D fricatives is not so complete as with plosives, and often the difference is more one of tenseness than of voicing. That is, <v> in onset is often devoiced, and indicated in some sources as [v], which is unvoiced and lax, while [f] is unvoiced and tense. The feature lax/tense differentiates the outer <v> and <f> consonants from one another in words like vijf and verf. The [??] devoicing also serves to differentiate words beginning with <v> and <w>, as in val/wal, vacht/wacht, vilt/wild, and vorst/worst. English <w> and <v> can thus be expected to create challenges for a D speaker, as for a German.

Dutch [c], [n], [??], and [3] are allophonic variants of [t], [n], [s], and [z], respectively, occurring only before [j]. The common E [??] thus occurs in many more environments than D [??]. Unlike Italian, intervocalic <s> becomes [z] only after long vowels.

Double consonants, like in G, are pronounced the same as single consonants, and usually indicate that the preceding vowel is short. D also shares with G the devoicing of voiced plosives in coda position:

[b] [right arrow] [p] heb, slab

[d] [right arrow] [t] hoed, raad

Thus, noot and nood are homophones. D speakers must be careful to differentiate in E between minimal pairs such as rate/raid, right/ride, sat/sad, rip/rib, and slap/slab. The same can be said of [f]/[v], so that life and live are not homophones.

Dutch <ch> and <g> are both pronounced [x]. Thus, juichen and duigen are exact rhymes, as are vlaggen and lachen. Lagen however has long [a:]. E words beginning with <g> will present difficulty, since neither [g] nor [d[??]] is employed in that environment in D, but rather [x]. The <sch> common to D, E, and G has different pronunciations in each, which can be a source of confusion. English school looks like Dutch school [sxo:l], and German schon looks like Dutch schoen [sxu:n]. Dutch <schr> omits the <ch>, as in schreeuw [sre: u]. Similarly, D <isch> becomes D [is], unlike G [i[??]].

In standard Dutch, /[??]/ is [R] or /[??]/ in onset (raam, droog), but [??] in coda (borst). (10) In coda, clusters beginning with <r>, as in borst, are pronounced in a remarkably similar manner to American English, and unlike German. Thus, D staart sounds much like E start, and unlike the [x] one hears in G warten in speech. Onset <r> in English [??] thus will have to become routine for a Dutch speaker, as [R] or [??] will be the reflex articulation. This is a good example of an allophone that is present in the home language, but is distributed differently in English. It is the context that must be acquired, not the articulation itself. In keeping with other languages, the D /[??]/ phoneme exhibits wide dialect and individual variation. D consonant charts will vary according to how much information they wish to convey, and which form of D is being outlined. As we have seen, D /[??]/ is realized as a uvular or an alveolar consonant, but trill, fricative and approximant allophones are used in both environments by different speakers and in different regions. Only the most common standard forms are indicated in Figure 1. Vocalic forms of /??]/ are also encountered, as in E and G. Uvular articulations have gained ground over the past several decades in speech, but the best strategy for lyric diction is to restrict the allophones to alveolar and vocalic articulations, as in G singing. Some coda /[??]/ clusters insert an epenthetic schwa, as in kerk ['ke.[??][??]k], arm ['a.[??][??]m], erf ['e.[??][??]f], erg ['e.[??][??]x], and dorp ['d[??].[??][??]p]. This is a consideration of the spoken tongue, and does not apply in singing. (11)

The foregoing contrastive discussion does not pretend to be a thorough treatment of the Dutch sound system, and is intended only to point out how a contrastive analysis can assist in anticipating problems that might arise in the voice studio. Note that much of the most useful information for contrasting E and D cannot be derived simply from step one (contrasting tables), but must involve access to sources that describe steps two and three in some detail. For instance, at the suprasegmental level, Dutch has a set of phrasal assimilation rules that are important to understand and would only be encountered in step three.


A primary purpose of this article is to encourage the voice instructor to delve into what may be uncharted waters, with a view to equipping oneself with basic tools of understanding the sound patterns of unfamiliar languages that a student may possess from family background and/or life experience. This allows the instructor to:

1. anticipate potential inaccurate phonetic transferences from that language to English, or to other languages of vocal repertory;

2. recognize readily those spots, and understand why they are occurring;

3. explain to the student not only the error, and that it must be corrected, but also why it is occurring, with reference to what is familiar to the student-perhaps even with specific examples of words from the student's native language. It has been the present author's experience that such students appreciate such contrasts much more readily, and with greater comprehension, when an error in unfamiliar territory for the student is compared to something with which they are thoroughly familiar.

This background work on the part of the pedagogue comes with a crucial caveat, however. Language acquisition aids come in a bewildering variety of forms and approaches, designed for different types of learners. Educational tools tend to congregate around two extremes. At one extreme are the chatty practical guides designed for those with no background in linguistic tools such as the IPA. These can range from travel guides with a one-page summary of how to pronounce a language ("like English UUHH"--whatever that means) to online resources that go into considerable detail, with helpful sound clips, but still assume that the reader has no familiarity with IPA. At the other extreme are academic monographs that presuppose a firm grasp of linguistic science (at least at the undergraduate level), and are not intended as practical guides. Like Cinderella in Sondheim's Into the Woods, what the voice instructor needs is "something in between"--a source with the precision of scholarly reporting, employing IPA as the basis of discussion, yet accessible in all other ways to the linguistic layman. Many online resources are marred by the attempt to make things comprehensible by employing supposedly familiar alphabetic renditions of sounds, such as are found in many older dictionaries. The problem is twofold: not only are such simplifications inherently imprecise, but there is also no standard to which all adhere. The reader becomes mired in uncertainty, or worse, misled into a false certainty. It is essential that the search for information about sound patterns in any language be done with a healthy dose of skepticism regarding sources of information, and an ability to separate the linguistic wheat from the chaff.

The thrust of this article has been towards contrasting languages with English only. Of course, English becomes merely the lingua franca with an ESL voice student in the studio whenever the repertory itself is in a different language. If a singer from Armenia is learning a German lied, the point of comparison now shifts to Armenian and German, with English taking a back seat. An instructor who is fluent in neither language can still assist in correct articulation by adopting the systematic approach outlined above for the languages involved-or for that matter, between any two languages. This may seem a presumptuous undertaking, especially if the student is fluent in one of the two languages and the instructor is not. There are certainly limits to what an instructor should feel comfortable doing in such unusual circumstances, but it must also be remembered that there is no guarantee that conversational fluency in any language will translate automatically into good sung diction, either from a textual or a technical point of view. Textually, the student may possess a strongly marked regional dialect inappropriate for the repertory, sometimes without realizing it. Technically, problems can surface in one language that become less of a concern or effectively disappear in another, for a variety of reasons.


Tables and Figures





TABLE 1. American English consonants.

                  Bilabial   Labiodental   Dental/Alveolar/Postalveolar

Plosive              Pb                                 td
Nasal                m                                  n
Flap                                                  ([??])
Fricative                        fv        [??][??]     sz     [??][??]
Approximant                                            [??]
Lateral approx.                                         l

                  Palatal   Velar   Uvular   Glottal

Plosive                      kg                 ?
Nasal                       [??]
Fricative                                       h
Approximant          j        w
Lateral approx.   ([??])


1. Affricates [t[??]], [d[??]], not included in Table 1,
are considered separate phonemic units by some.

2. [r], [[??]], and [[??]] are allophones of /r/, with
[[??]] the predominant form in speech.

3. The velarized [[??]] found in syllable codas is an
allophone of /l/.

TABLE 2. French consonants.

                  Bilabial   Labiodental   Dental/Alveolar/Postalveolar

Plosive           pb                       td
Nasal             m                        n
Trill                                               (r)
Flap                                                ([??])
Fricative                    f v                    sz       [??][??]
Lateral approx.                            l

                  Palatal   Velar    Uvular   Glottal

Plosive                     kg
Nasal             [??]      ([??])
Fricative                            [??]
Approximant       j,[??]    w
Lateral approx.


1. The trill and flap /r/s are sung articulations, not found in
standard spoken French.

2. The velar ([??]) is not an articulation of French, but is
employed in foreign borrowings.

TABLE 3. Dutch consonants.

                  Bilabial   Labiodental   Dental/Alveolar/Postalveolar

Plosive           pb                               td       (c)
Nasal             m                                n
Flap                                               [??]
Fricative                    fv                    sz       [??][??]
Approximant                                        ([??])
Lateral approx.                                    l

                  Palatal   Velar   Uvular   Glottal

Plosive                     k                ?
Nasal             ([??])    [??]
Trill                               (R)
Fricative                   xy               [??]
Approximant       j
Lateral approx.


1. Dutch has no affricates.



Brown, Keith, and Sarah Ogilvie. Concise Encyclopedia of Languages of the World. Oxford: Elsevier, 2009.

Campbell, George L. Concise Compendium of the Worlds Languages. London/New York: Routledge, 1995.

Comrie, Bernard, ed. The World's Major Languages. New York: Oxford University Press, 1987.

Delattre, Pierre, and Bertil Malmberg. Studies in Comparative Phonetics: English, German, Spanish, and French. Heidelbert: Gross, 1981.

Fisiak, Jacek, ed. Contrastive Linguistics and the Language Teacher (Language Teaching Methodology Series). Oxford: Pergamon Press, 1981.

Katzner, Kenneth. The Languages of the World, 3rd ed. London/New York: Routledge, 2002.

Kellerman, Eric, and Michael Sharwood Smith. Crosslinguistic Influence in Second Language Acquisition (Language Teaching Methodology Series). New York: Pergamon Institute of English, 1986.

Ruhlen, Merritt. A Guide to the World's Languages. Stanford, CA: Stanford University Press, 1987.

Walz, Joel. The Early Acquisition of a Second Language Phonology (Phonetic Reports from Hamburg: Investigations in Phonetics and Linguistics 28). Hamburg: Helmut Buske, 1979.


Dempsey, G. Anna. "Comparing the Vowels of English and French: A Contrastive Phonology." MA thesis, University of Illinois, 1976.

Tranel, Bernard. The Sounds of French: An Introduction. Cambridge: Cambridge University Press, 1978.

Vezina, Louise Rosealma. "The Measurement of French Pronunciation, Based on the Contrastive Analysis of Standard French and General American English." PhD dissertation, Georgetown University, 1975.


Aarts, Flor, and Herman Wekker. A Contrastive Grammar of English and Dutch, 2nd ed. The Hague: Martinus Nijhoff, 1993.

Booij, Geert. The Phonology of Dutch. Oxford: Oxford University Press, 1999.

Collins, Beverley, and Inger Mees. The Sounds of English and Dutch, 5th ed. Leiden/Boston: Brill, 2003.

Donaldson, B.C. Dutch: A Linguistic History of Holland and Belgium. Leiden: Martinus Nijhoff, 1983.

Hulst, Harry van der. Syllable Structure and Stress in Dutch (Linguistic Models 8). Dordrecht: Foris Publications, 1984.

Lagerwey, Walter. Speak Dutch: An Audio-lingual Course, 6th ed. Amsterdam: Meulenhoff Educatief, 1976.

Oostendorp, Marc van. Topics in the Phonology of Dutch. Amsterdam: Meertens Institut, 2006 (online resource).

Shetter, William Z. Introduction to Dutch. The Hague: Martinus Nijhoff, 1977.

Leslie De'Ath, Associate Editor


(1.) Robert P. Stockwell and J. Donald Bowen, The Sounds of English and Spanish (Chicago and London: University of Chicago Press, 1965), p.vii. The other volumes in the series are: William G. Moulton, The Sounds of English and German (1962); Frederick B. Agard and Robert G. Di Pietro, The Sounds of English and Italian (1965).

(2.) Don L.F. Nilsen and Alleen Pace Nilsen, Pronunciation Contrasts in English (New York: Regents Publishing, 1971).

(3.) Ibid., 12.

(4.) The present author demonstrates this circumstance between English and Korean in his article, "Linguistic Lingo and Lyric Diction I--The Phoneme," Journal of Singing 67, no. 2 (November/December 2010): 189.

(5.) Among published sources in book form, the author recommends the Brown/Ogilvie, Campbell and Comrie reference works cited in Related Literature below.

(6.) Bernard Tranel, The Sounds of French: An Introduction (Cambridge: Cambridge University Press, 1987), includes French-English contrastive study chapters throughout, with the focus on spoken French.

(7.) The status of [??] is controversial among phoneticians, and sometimes [h] is used for D too. It is sufficient to know that the <h> of D is often more tense and voiced than in E.

(8.) In D, [??] is always short, while [??] is always long and often diphthongized. In older sources D [??] is sometimes indicated as [??], the symbol for a high, central, rounded vowel, even though D [??] is considerably less closed.

(9.) When followed by [??], these diphthongs centralize: [e[??]], [[??][??]], [o[??]].

(10.) Unlike in E or G, the symbol chosen for the phoneme in standard D is /[??]/ rather than /[??]/, since it is the most prevalent allophone.

(11.) If we return now to the Nilsen book to check the lists of languages presumed to have difficulty with vowel and consonant contrasts within English, we find that Dutch is listed as a likely language in the following cases:
[ae] vs.   [[??]]        cat / cot
[h]        [[??]]        hack / whack
[w]                      [[??]] Wac / whack
[w]                      [[??]] Wac / rank
[v]                      [[??]] vat / that
[t]        [[??]]        tank / thank
[[??]]     [[??]]        thank / shank
[s]        [[??]]        sank / thank
[[??]]     [[??]]        thigh / thy
[d]        [[??]]        Dan / than
[z]        [[??]]        Zen / then
[t]        [d] (final)   tab / dab
[d]        [r]           dash / rash
[d]        [z]           dipper / zipper
[l]        [[??]]        lack / rack

The model words are those of the Nilsen book. The reader may
wish to compare these contrasts with those earmarked in
this article.

Example 1.

[i(:)]      i, ie                         prima, zie, dierbaar
[i]         i                             wind, zitten
[y(-)]      U, UU                         nu, duurzaam, student
[[??]]      U                             lucht, nut
[[??](:)]   e                             en, perzik, respect, zet
[a:]        aa, a (open syllable)         ja, daarom, natie
[[??]]      a (closed syllable, or        badpak, extra, nat
              open unstressed syllable)
[[??](:)]   o                             donder, zot
[u(:)]      oe                            genoeg, hoed
[[??]]                                    behang, geschenk
[e:]        e, ee                         (= [e:i]) serie, beet
[[??]:]     eu                            (= [[??]:y]) neus, terreur
[o:]        oo                            (= [o:u]) boot, code, zo
[[??]i]     ij, ei                        dijk, geheim, mei
[oey]       ui                            lui, ui
[[??]u]     ou, ouw, au, auw              nauw, nou, gebouw
COPYRIGHT 2011 National Association of Teachers of Singing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:De'Ath, Leslie
Publication:Journal of Singing
Date:Nov 1, 2011
Previous Article:The economy of choir size.
Next Article:Pedagogic pears: exploring vocal resonance.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |