By EDWARD ARTIN , Pronunciation Editor
This is a presentation of facts adequate, it is hoped, to explain the way we use our symbols in pronunciation indications in the vocabulary and to enable the consultant to infer certain pronunciations not actually shown from pronunciations that we do show. This is not a treatise on phonetics; many elementary facts of phonetics are not discussed at all (for example, the nature of most articulations). Those who desire such information should consult any of a number of good textbooks on phonetics.
The matter in this “Guide” is arranged under headings that so far as possible are in alphabetical order. Criteria for the alphabetization of some of the items have not been established by usage.
Headings that are not, that do not contain, or that are not formed by addition to, the 26 letters of the English alphabet appear ahead of everything else. Between the beginning of the first of such items and the end of the last no scheme of ordering is attempted.
The character \ə\ and modifications of it appear next.
The character \ŋ\ appears between the n, \n\ items and the o, \ō\, \ȯ\ items.
Ligatured characters are alphabetized as if they were separate letters (e.g., \œ\ is alphabetized as if it were \oe\).
A character entered as an orthographic letter appears ahead of the same character entered as a symbol of our pronunciation alphabet.
The same symbol may appear in two or more of the following forms: plain or unmodified; with diacritic above or below; reduced in size and set high as a superior character. For a symbol having two or more of these forms the alphabetization is as follows:
(1) single plain symbol, as \a\, \th\
(2) single symbol with horizontal bar above or below, as \ā\, \th \
(3) single symbol with two dots, as \ä\
(4) single symbol with one dot, as \ȧ\
(5) single symbol set as a small superior, thus: \ə\, \ə̄\, \ə̇\, \ə \
(6) multiple-character symbols come at the end of the last single symbol based on the first character of the multiple symbol. Thus \aa\ comes at the end of the series \a\, \ā\ \ä\, \ȧ\, not between \a\ and \ā\. Such multiple symbols are alphabetized without regard to diacritics or to the size of any component character, except that when such disregard results in duplication the order is as stated in (1)–(5) above. Thus \ai\ precedes \äi\ and \ōō\ precedes \ȯȯ\
\ \
see the section on phonemicity .
( )
two parentheses mean that whatever is indicated by the symbol or symbols between them is present in the pronunciation of some speakers and absent from the pronunciation of other speakers, or that it is present in some utterances and absent from other utterances of the same speaker, or that its presence or absence is something difficult to be sure of.
nu·mer·ous \ˈn(y)üm(ə)rəs\
floc·cu·lence \ˈfläkyələn(t)s\
chick·a·dee \ˈchikə(ˌ)dē, …\
Such pronunciations could alternatively have been shown, at greater cost of space, as \ˈnümərəs, ˈnyü-, ‑mrəs\, etc.
Parentheses are often placed around one of the members of a digraph symbol when there are two pronunciation variants one of which is symbolized by the digraph and the other of which is symbolized by one member of the digraph.
var·i·ous \…, ˈva(a)r-, …\ [\ˈva(a)r-\ = \ˈvar-, ˈvaar-, or vice versa \]
su·mac \ ˈs(h)üˌmak, …\ [\ˈshü-\ = \ˈshü-, ˈsü-, or vice versa \]
fair \ˈfa(a)(ə)r, …\ [\ˈfa(a)(ə)r\ = \ˈfaər, ˈfar, ˈfaaər, ˈfaar, or any other order \]
There will possibly be those who will object to such symbolization on the ground that, for example, \sh\ is a single sound, not two, that \sh\ really has no \h\ in it, and that it is bad to suggest that \sh\ can be converted to \s\ by omitting a final \h\ sound. Our view is that \s(h)\ is no more objectionable than \sh\, and that dictionary users who demand \sh\ should be prepared to accept \s(h)\.
In an abbreviated variant pronunciation following a punctuation mark one of two parentheses in a preceding pronunciation may be repeated simply to provide a locator for the substitution of the second variant in a preceding pronunciation.
cre·do \ˈkrē(ˌ)dō, ˈkrā(-\ [the \ā\ replaces \ē\ between \ˈkr\ and the first parenthesis]
An r that stands at the end of a transcription will often be found parenthesized, and the parenthesizing takes either of two forms: a left-handed parenthesis stands before the r and no right-handed parenthesis follows, as in \ˈmär, ˈmȧ(r\ at ¹mar; a parenthesis stands on either side as in \ˈbäth ə(r)\ at ¹bother. The single parenthesis will be found only in a pronunciation in which there are at least two variants separated by a comma, and never in the first variant. A preceding variant ending in r without parenthesis shows the pronunciation of those who have as many \r\ sounds in their speech as there are letters r and letter sequences rr in the spelling (but see the section on dissimilation ). A variant with a single parenthesis preceding the r shows the pronunciation of those who often do not have or who never have immediately before a consonant or a pause the sound, made usually by a lifting and curling backward of the tip of the tongue, that is transcribed \r\. For details of the environments in which \r\ is pronounced and those in which it is not, see the section on \r\ .
A transcription with a parenthesis on each side of a final r, as \ˈbäth ə(r)\ at bother, is an abbreviated way of writing the transcription \ˈbäth ər, ˈbäth ə(r, with explanation as for \ˈmär, ˈmȧ(r\ in the preceding paragraph. The abbreviated transcription can be used for bother but not for mar for the reason that the vowel symbol immediately preceding the r is the same for both kinds of speech in bother but not in mar. Since by definition above the single parenthesis means that r occurs in some environments but usually not in others, we could not use \ˈbäth ə(r\ alone as our transcription, which would not hold for those who have r in all environments. Our explanation of double parenthesis, however, in the first sentence of this section makes the two-parenthesis transcription applicable to both r ‑keepers and r ‑droppers.
=
see 2.7.1, 2.7.2 in “Explanatory Notes ”.
+
see 2.9 of “Explanatory Notes ”, and stress marks and “…” in this “Guide”.
When a stress marking (ˈ) or ¦ appears on a syllable in the item that precedes the plus sign, the marking ¦ is to be understood as superseding a primary stress in the pronunciation that is to be supplied for the last part of the compound.
The plus sign is not necessarily an indicator of the etymological make-up of the word in whose pronunciation it is used. The etymology in square brackets, immediately following the pronunciation, is the proper source of such information. Thus the etymology of su·per·naturalism is plainly shown as [¹supernatural + ‑ism ] but as a time-saving and space-saving device the pronunciation is shown as \ˌsüpə(r)+\ (that is, “+ the pronunciation at naturalism ”).
When the spelling that the plus sign directs the consultant to refer to in order to complete the pronunciation has two or more entries belonging to different parts of speech not identical in pronunciation, the pronunciation to be supplied is that of the spelling whose part of speech is the same as that of the compound. Thus for ¹concorporate, an adjective, add the pronunciation given for the adjective corporate, and for ²concorporate, a verb, add the pronunciation given for the verb corporate. If the compound has inflected forms for which the pronunciation is desired, this information should be sought at the entry for the word that is the last part of the compound.
When the second syllable of a word has primary stress, anything less than primary stress is not marked on the first syllable (except when the vowel of the first syllable is \ə\), although first syllables often have a degree of stress that we mark as secondary stress on any syllable except the first. Thus no stress mark appears before the \mī-\ at microscopist and microscopy, although ˌ would appear before the \-mī-\ of biomicroscopy if we were transcribing that compound in full. We usually forgo such nicety of stress indication when we use the plus formula, however, and the pronunciation of biomicroscopy is shown as \“ + \ (read ¦bīō for the ditto).
The reason for the use of \“ + \ instead of \“\ alone as the pronunciation of a word for which the value of the ditto is to be picked up from a preceding entry is illustrated by the following sequence of entries:
neo·arsphenamine \¦nē(ˌ)ō+\
neo-assyrian \“ + \
¹neo-babylonian \“+\
²neo-babylonian \“\
Use of \“\ alone at neo-assyrian and ¹neo-babylonian could mean that these two words have the same pronunciation as neoarsphenamine. Use of \“\ alone at ²neo-babylonian means that that word does have the same pronunciation as the entry preceding, ¹neo-babylonian.
When an initial combining form has variants and appears in a number of compounds, we often list the variants in the entry for the combining form alone, devise a formula that we equate to the variants, and pronounce compounds in which the combining form so pronounced appears by means of the formula and a plus sign.
elec·tro- \in pronunciations below, =¦== = ə̇¦lek(ˌ)trō or ē¦- or ‑- trə\
…
¹electrodeposit \=¦== at electro- +\
The principal stress in the combining-form entry is altered if the stress in the compound differs.
In entries pronounced with the plus formula, usually no division marks are shown in the second part. This part may be divided at the places indicated at the entry where the pronunciation is to be sought. Divisions between the beginning of the compound and the beginning of the second element are usually shown, except that in compounds like ¹electrodeposit both the pronunciation and the divisions for the first part are to be sought at the entry for the initial combining form. In such compounds a division may also be made between the two components; in fact, this division is preferable to any other.
Many printers avoid dividing after a single initial letter, and no division is shown between the components of a compound when the first is a single letter, as in asexual.
|
the vertical bar, ascender-high and descender-deep so that it will be as distinct as possible from \l\, is frequently used in transcriptions in which one or more variants are shown as an orientation mark to make it easier for the consultant to locate the part of the pronunciation in which the variation occurs. Our usual method of fixing the point of variation is to repeat, to both the left and the right of the variation or to one side only, one or more characters that are common to the variants and that fulfill other functions in the transcription (sound symbol, stress mark, parenthesis). But when contiguous variations occur, or when a character that would ordinarily be used occurs more than once in a transcription, a character whose sole function is that of a locator is desirable. The vertical locator is inserted in the first variant and is repeated in the same position in one or more variants that follow:
flight·i·ly \ˈflīd·|ə lē, ‑īt|, |ə li, |ə̇l-\. See 2.6 in “Explanatory Notes ”.
Two vertical bars have been used with some frequency to facilitate the presentation of three contiguous variations, as in
gen·er·a·tive \ˈjenəˌrā|d·|iv, ˈjen(ə)rə|, |t|, |ēv also |əv \
In the variants presented after the first mark of punctuation variants of the first of the three contiguous parts end with a vertical bar, variants of the second have a vertical bar on each side, and variants of the third begin with a vertical bar.
“
see 2.8.1, 2.8.2 in “Explanatory Notes ”.
·
We hear a difference between the usual American pronunciation of the two items fly twitch and flight witch and the two items sent reagents and sentry agents, for the members of each of which pairs most transcribers would show identical sequences of sound symbols. On the other hand, we do not hear a difference between the two items wool fox and wolf ox, nor between the two items wool freight and wolf rate, for the members of each of which pairs most transcribers would also show identical sequences of sound symbols. (The qualification is to be understood that our observations are based upon the pronunciation of all items without pause, punctuative or hesitative. That some are nonsense items is of no moment.) We conclude therefore that an alteration of the boundary between meaningful elements in a sequence, or that the introduction of such a boundary into a sequence that lacks it, can cause a phonetic difference in some sequences but not in others. It makes sense to treat this phonetic difference as a difference of syllable division rather than as a difference between certain sounds in the pairs of sequences, which actually it is: such an interpretation will be more readily understood by those who do not know phonetics, and a single symbol for syllable division will do work that alternatively requires several additional sound symbols. Our symbol for a syllable division is a centered period, ·; we would use it in one position in a full transcription of items like fly twitch and sentry agents (\-ī·tw-\, \-n·tr-\) and in another position in items like flight witch and sent reagents (\-īt·w-\, \-nt·r-\); we would not use it at all in a full transcription of any of the items wool fox, wolf ox, wool freight, wolf rate. (Actually, we usually omit it in items of the first group when they are compounds of two English words and the pronunciation is limited to stress indication.) Our stress marks are not simultaneously indicators of syllable division, as they are in some systems of transcription, and their placement when no · accompanies them is arbitrary Thus our transcription at astronomer is \əˈstränəmə(r)\ but it is impossible to know from a comparison of normally pronounced English items containing the sequence \st\ whether the syllable division is before the \s\, between the \s\ and \t\, or after the \t\. In short, our policy in this book is to indicate a syllable division (use the mark ·) only in sequences.
Examples of our use of ·:
be·tween \bə̇·ˈtwēn …\ [\tw\ different from that of rabbit warren. ]
sight \… usu ‑īd· + V\ [The voiced sound that most Americans have between the first two vowels of sight over is different from the voiced sound that some speakers (e.g., some British speakers) have between the first two vowels of side over or Cy Dover. ]
bur·ry \ˈbər·|ē …\ …: abounding in or containing burs … [Many have \ˈbər·ē\ for this but \ˈhə·rē\ for hurry. ]
alma-ata \¦almə·ə¦tä\ [The centered period is necessary here because in our system a doubled symbol with nothing between indicates a long monosyllabic vowel.]
For \r(·)\ at the end of a transcription, as at knur, see the section on \r(·)\ .
-
the low-set minus sign usually cancels a stress at the same point in a pronunciation shown elsewhere, as in a preceding variant within the same reversed virgules or in a pronunciation for a preceding word in the same or another entry.
cam·era·man \ˈ=(=)=ˌman, ‑- mən\
¹mer·cu·ri·al \¦mər¦kyu̇rēəl, - mə(r)ˈk-, …\
ac·tiv·ism \ˈaktə̇ˌvizəm\
¹ac·tiv·ist \-- və̇st\
Occasionally - merely emphasizes that the syllable that follows is without stress, as when it stands at the beginning of the first pronunciation shown for a monosyllable.
from \- f(r)əm, …\
¹‑y also ‑ey \- ē, i\
If the compound word binaural had been transcribed in full in the vocabulary, the stress indication would have been ¦ and would have been placed immediately ahead of the symbol \n\. In some areas of the vocabulary we used a formula of the type \(ˈ)bī¦n+- -\ for such compounds (see the plus sign ), in which the purpose of the mark - is to cancel the stress appearing before the first vowel at the entry aural in favor of the placement of stress indication for the same syllable before the \n\ in the compound, and so to guard against the interpretation that our intention as to stress is \(ˈ)bī¦nˈȯrəl\. However, since we do not employ stress marks to double as syllable dividers, formulas of the type \(ˈ)bī¦n+- -\ may carry nicety of stress-mark placement to excess. The shift of stress indication for au from ˈ at aural to ¦ at binaural is implied by (ˈ) or ¦ in the first member of the compound, as it of necessity must be when the plus formula is used for a compound whose second member has no stress on the first syllable. Hence in most of the vocabulary we use a formula of the type found at exostracize, \eks+ \ (instead of \ekˈs+- -\).
÷
see 2.12 in “Explanatory Notes ”.
…
the three-dot ellipsis is often used between sound symbols preceding and following for a pronounced part that will be found in a corresponding position in a preceding pronunciation either within the same or in a preceding entry.
humpty-dumpty \¦həm(p)tē¦dəm(p)tē, ‑ti … ti\ It may be so used for one or more of the following reasons: what it stands for may be of such length that to repeat it would consume space and distract the reader's attention from the part of the word to which we wish to draw it; it may emphasize that the pronounced parts preceding and following are concomitant variants; a duplication of sound symbols in the part of the pronunciation the ellipsis stands for may make the provision of orienting symbols from this part difficult. For additional examples see dilettantist and catalogue raisonné in the vocabulary.
Occasionally … is used at the end of a pronunciation consisting of stress indication only for a multiplicity of variants that it would be clumsy to indicate otherwise.
high-temperature \ˈ=¦= …\ [see temperature ]
In a formula such as that found in the pronunciation of the three-part entry
sul·fo·benzoic acid \¦səl(ˌ)fō + … ‑\
the three dots stand for the second of the three parts. The dots together with the hyphen that follows tell the consultant that to find the pronunciation of the entire entry he must look at two other entries, benzoic or benzoic acid, and acid.
\ə\
when a stress mark (ˈ or ˌ) stands at the beginning of the syllable in which it occurs, this symbol, called schwa, is pronounced as in bud or nut or the last syllable of aqueduct; when the syllable in which it stands is without stress mark, it is pronounced as in the first syllable of alone or occur or as in the second syllable of colony or as in the last syllable of abbot or famous or sabbath or circus.
Formerly nearly all phonetic alphabets used for the vowel of bud a symbol different from that for the vowel of the second syllable of abbot, and some alphabets still do. Some who are familiar only with these alphabets find the use of \ə\ in stressed syllables objectionable when they encounter it for the first time. But use of \ə\ as a symbol for both unstressed and stressed vowel is rapidly increasing, and abandonment of a separate symbol for the vowel of bud parallels abandonment of former symbols for half-long a, e, and o in whose stead ā, ē, and ō without stress mark are entirely adequate.
\ə̄\
used by r ‑droppers for the ‑ir of stir and the ‑irre- of stirred, and used by many r ‑droppers for the ‑ur- of sturdy and the ‑ir- of bird. It is a single sound and not a diphthong.
An r ‑dropper, symbolized by −R in transcriptions in this dictionary, in no case makes immediately before a pause or a consonant the r ‑sound that is characterized by the raising and bending backward of the tongue tip. Complete r ‑droppers are comparatively rare, and in the U.S. possibly nonexistent. A not uncommon phenomenon is a speaker who pronounces bird and the first syllable of further the way an r ‑keeper does but who drops the r of the second syllable of further and of words like farm and form.
In words like bird and sturdy (words not having r as the last letter in the orthography and not derived from such words) but not in words like stir, stirred, stirs, many r ‑droppers in the New York City area and in the southeastern U.S. have instead of the vowel \ə̄\ the diphthong \əi\ (see the section on \əi\ ). A variant vowel used by some speakers for words like bird and stir but not shown in this dictionary is \əə\ — that is, \ˈə\, the vowel of bud, lengthened. With such speakers bud and bird have the same vowel, but it is appreciably longer in bird. With r ‑droppers in the U.S. the pronunciation of words like furry and stirring (two-element words whose second element is a vowel suffix or a vowel-initial suffix) is either \ˈfə̄rē\, ˈstə̄riŋ\ or more often \ˈfər·ē\, \ˈstər·iŋ\, the latter being also the pronunciation of r ‑keepers. See the section on \ər(·)\.
\ə̄\ is often used by r ‑droppers as an anglicization of any of the vowels \œ\, \œ̄\, \œœ\, \ə̄ə̄ in words borrowed from a foreign language. Example: \-sə̄z\ for the second syllable of berceuse which in French is \-sœ̄œ̄z\. This \ə̄\ anglicization is even used to some extent by r ‑keepers although it is not a sound that they use in fully English words. See \œ\.
\ə̇\
used in unstressed syllables only; the dot is not a diacritic indicating that \ə̇\ stands for something different from \ə\. Rather \ə̇\ is a compound or two-part symbol, the two components being \ə\ and the dot of the symbol \i\. \ə̇\ is used when \ə\ occurs in some dialects and \i\ in others, or when the same speaker may have \ə\ in ordinary speech but \i\ in formal speech, or in an initial position where either \ə\ or \i\ may occur depending on what precedes. Less space is required to transcribe, for instance, rapid as \ˈrapə̇d\ than to transcribe it as \ˈrapəd, ‑pid\. In positions in which we show \ə̇\, most dictionaries have long shown only \i\ — or even \e\ alternatively or alone when the orthography has e, as in the second syllable of ticket. is the pronunciation of some Americans and of most speakers of Received Standard, but most Americans make exact rhymes of quotaed and quoted, of ballad and valid, of abbot and rabbit and rabbet. The first member of each of these groups is shown with only \ə\ or its equivalent as the unstressed vowel.
Our system includes the three symbols \ə\, \i\, and \ə̇\ and uses \ə̇\ to mean either \ə\ or \i\. A system (or a modification of it) used by many linguists includes the three symbols \ə\, \i\, and “barred i ”, produced by overprinting the \i\ with a hyphen), and uses barred i in most of the places where we use \ə̇\. As used by most, it appears to be intended as a symbol for a sound distinct from \ə\ and \i\ — articulated with the tongue higher than for \ə\ and farther back than for \i\. Although it is quite true that the unstressed sound we symbolize \ə\ varies in its articulation more than any other vowel, the variation seems purely a matter of phonetic environment. If speakers do not consistently articulate within the same environment three distinct vowels symbolized by \ə\ and \i\ and barred i, then we need no equivalent for an unstressed barred i.
Barred i has also been used to symbolize a stressed vowel heard from some Southern speakers in the first syllable of sister, produced by drawing the tongue back in the direction of \u̇\. We have not used a symbol for this vowel in the vocabulary because the geographical and lexical incidence of the vowel does not seem to be well enough known. Some students of American speech insist that a symbol for such a stressed vowel is necessary for adequate transcription of the full range of variants heard for the adverb just from speakers in all parts of the U.S., on the ground that this adverb frequently has a vowel that is not the vowel of the adjective just, not the vowel of the noun gist, and not the vowel of the noun or verb jest. We agree but we are not convinced that (outside those areas of the South where the first vowel of sister is not \i\) pronunciations of the adverb just when the vowel is not stressed \ə\, \i\, or \e\ have any more stress to them than does the pronunciation of the gest of largest. The full range of variants for the adverb just is reasonably well covered for the English-speaking world as a whole when it is indicated that the vowel is \ə\, \i\, or (less frequent than the other two in standard speech \e\, all either stressed or unstressed.
In transcriptions of uncommon words, as a space-saving device, \ə̇\ only is used where \ē\ is heard as a variant of \ə\ and \i\ as in the antepenultimate syllable of words in ‑ical or in the second syllable of Libyco-Berber. Such practice is not inconsistent with the widely followed practice of transcribing as \i\ the last vowel of words like happy, a vowel for which the symbol \ē\ better represents the usual U.S. pronunciation. The \ē\ variant is usually included in transcriptions of common words.
If we were attempting to cover British as well as American pronunciation, we would use \ə̇\ much more freely than we do in vocabulary pronunciations. In British usage an unstressed vowel whose orthographic counterpart is i or e is usually \i\, but American usage more often has \ə\ than \i\ in unstressed syllables. Even with American speakers who most often have \i\, its range is much more restricted.
\ə \
(preceding \l\, \n\, \m\, \ŋ\; for ə following \l\, \m\, \r\, see \lə \ , \mə \ , \rə \ ) printed as a superior character means that a consonant following it is a syllabic consonant, that is, a consonant that immediately follows another consonant without any vowel between. Nearly all phoneticians are agreed that \n\ and \l\ are often syllabic in English although some use no symbol that indicates syllabicity. Some transcribers of pronunciations in dictionaries who use a special symbol for syllabicity with \n\ and \l\ find no occasion to attach the symbol to \m\ or \ŋ\, but this does not necessarily mean that they deny the existence of syllabic \m\ and \ŋ\. There is variation in the interpretation and transcription of the sound of the er in the three-syllabled pronunciation of coppery and in r ‑keepers' pronunciation of copper, some phoneticians regarding these sounds as a syllabic \r\.
Apparently there are many, especially among writers of textbooks on phonics and teachers in the elementary schools, who are of the opinion that there can be no syllable without a vowel. Every dictionary in line of succession from, and including, Noah Webster's original dictionary of 1828 shows vowelless syllables. For instance, the unabridged Webster dictionaries that preceded this pronounce the word kitten as follows:
1828 kit′n
1847 kit′tn
1864 kĭt′tn
1890 kĭt′t'n
1909 kĭt′'n
1934 kĭt′'n
As a realistic representation of what takes place in articulation when a syllabic consonant is pronounced, the representation employed by Noah Webster in 1828 is difficult to improve: the consonant alone is used without the addition of any symbol for syllabicity. Something additional must be used when the syllabic consonant is not final in a sequence and is followed by a vowel; thus \kitnish\ does not adequately represent a three-syllabled pronunciation of kittenish, but its inadequacy can be rectified by inserting a hyphen after the syllabic consonant, thus: \kitn-ish\). Our latest school dictionaries transcribe kitten as \ˈkit-n\, and with the abolition of unnecessary marks of syllable division in this book our transcription could have been \ˈkitn\. However, this book departs from this method of transcribing syllabic consonants for two reasons:
(1) In some environments some monosyllabic particles (for example, and ) often have no vowel sound but have a syllabic consonant as their first or only sound. Although in the transcription of words of more than one syllable a special symbol for syllabicity is not essential, without one the indication of a syllabic consonant in a monosyllable is a matter of some difficulty, and in this book the user has a right to expect a minute listing of all the variants commonly heard for such monosyllabic particles as and.
(2) In our school dictionaries, a syllabic consonant followed in the same word by a vowel is indicated as syllabic by the printing of a hyphen (in these books a mark of syllable division) after the consonant, as in \ˈfas-n-ˌāt\ for fascinate and \ˌfas-n-ˈāsh-n\ for fascination. We have found that these transcriptions have been often understood to show fascinate as having two syllables only or fascination three syllables only, so we attempt to forestall misinterpretations such as these by using a small superior schwa before the syllabic consonant.
Some transcribers symbolize a syllabic consonant by a full-size \ə\ followed by the symbol for the consonant, for example, the sound that follows the second consonant of wooden being transcribed \ən\. Either or both of two lines of reasoning seem to motivate this procedure:
(1) There are speakers who have a vowel between the \d\ and \n\ of words like wooden. Even those who rarely or never have the vowel-plus-\n\ pronunciation would not apprehend that pronunciation as anything but the word wooden. There are few if any speakers who consistently have \d\ plus syllabic \n\ in one kind of context and have \d\ plus \ə\ plus \n\ in another context, and therefore the two pronunciations are seldom in contrast. Why then go to the trouble of having an additional method of symbolization for a syllabic consonant?
(2) The pronunciation without any vowel in the second syllable of wooden is the usual one, but a speaker's ingrained pronunciation habits can be depended upon to make him pronounce a syllabic consonant in the right places even though the transcription shows \ə\ before the consonant.
Against these arguments several counter arguments can be posed:
In the pronunciation heard, with few exceptions, for a word like wooden from speakers who are communicating with friends and are not trying to impress, there is no vowel in the syllable that contains the \n\. Some speakers do insert the vowel \ə\ before the consonant especially when they have a large audience containing hearers who may expect a more formal kind of pronunciation than the speaker ordinarily uses. That there are others in any such audience, however, who do not expect the speaker to alter his normal style of speech to this extent must be the experience of many or most who have done a great deal of radio or television listening in the company of others. When a television news broadcaster pronounces threaten with a vowel in the second syllable, a listener, although ignorant of phonetics, must be instantly aware that the pronunciation differs from what years of hearing English spoken have shown to be the usual pronunciation of threaten. Most speakers of English who pronounce a vowel in the second syllable of wooden or threaten probably do so because the spelling contains a vowel letter before the n. These speakers make a literal-minded interpretation of the presence of the written vowel, and do not see that they have introduced a vowel sound unnaturally. They can point to transcriptions that represent the same succession of symbols — \tən\ — the initial sounds of tonight and the final sounds of threaten. The utterances lye denotes and Leyden oats differ in that the first would ordinarily be pronounced with a vowel between the \d\ and the \n\ (with American speakers this vowel would often or usually be \ə\) whereas the second would ordinarily not have a vowel between the \d\ and the \n\. Some linguists would assert that there would be a difference of juncture between these two items whether there is or is not a vowel \ə\ in the second syllable of the second item — that in spite of identity of sounds and stresses and intonation there is something in the way these otherwise identical items are uttered that makes it clear to a hearer that the word boundaries are not the same. However, as readers know who are familiar with our vocabulary pronunciations or who have read the section on juncture in this “Guide”, it is our view that often it is impossible to identify word boundaries phonetically. The difference between lye denotes consistently pronounced with \ən\ after the \d\ and Leyden oats consistently pronounced with a syllabic \n\ after the \d\ is the only one consistently made. Such a consistently made difference is phonemic.
Transcribers who use the symbol \ə\ in the second syllable of words like wooden will apparently be in difficulty if they decide to record certain widespread pronunciations that all of them seem to have thus far avoided — the pronunciation of seven, for instance, in which the second consonant is \b\ rather than \v\ and the last consonant is \m\ rather than \n\. The consultant who interprets a transcription like \ˈthretən\ literally and pronounces the word with a vowel in the last syllable will be saying something that may be heard from numerous other speakers but the consultant who interprets a transcription like \ˈsebəm\ in the same way would be using a pronunciation almost without precedent.
There can be no better way of indicating no vowel between two consonants than to print them with nothing between them — e.g., to transcribe \ˈsebm\ (one pronunciation of seven ) and \ˈkäntn-ənt\ (one pronunciation of continent ), or to follow the long-established IPA practice of using a modifier underneath the symbol for the syllabic consonant, at least whenever the syllabic consonant is followed by an unstressed vowel in the same word. Some dictionary users, however, prefer some sort of symbol between the two consonants, and the raised symbol \ə \ should provide an acceptable compromise between the extreme of no modifier at all and the extreme of a full-sized \ə\, which tends to encourage over-pronunciation.
Transcribers of dictionary pronunciations who employ a special way of indicating that a consonant is syllabic treat \n\ and \l\ as syllabic in a high percentage of cases. Some dictionaries show no syllabic \m\ or \ŋ\ in any of their pronunciations, although the very best speakers from time to time use a syllabic \m\ and less often a syllabic \ŋ\. (On these, see the entries \ə m\ and \ə ŋ\ .) The phonetic entity that corresponds to the er of the spelling differ or to the or of the spelling honor in the speech of r ‑keepers, and often in the speech of r ‑droppers when a word beginning with a vowel sound follows without pause, is variously interpreted. Some phoneticians regard this entity as the vowel \ə\ pronounced with a raising or turning back (retroflexion) of the tip of the tongue and use for it the symbol \ə\ with a symbol for retroflexion attached. Other phoneticians regard the sound as a syllabic \r\ and use a transcription that parallels that for syllabic \n\ or \l\. Still others transcribe \ər\. Of this number some regard the item as two sounds, others are willing to grant its singleness but find the two-symbol transcription more practicable since it more closely parallels the orthographic spelling, does not require the additional specially made symbol that use of a single symbol might, and makes it possible to show both the r ‑keepers' and the r ‑droppers' pronunciation by merely parenthesizing or italicizing the r. Whereas the latest school dictionaries in the Merriam-Webster series transcribe the er of wither as a syllabic r — \ˈwith -r\ — this dictionary transcribes the same word \with ə(r)\. In the school dictionaries only the pronunciation of r ‑keepers is shown and it seemed advisable to attempt to indicate syllable boundaries in pronunciations (attempt because the exact location of many syllable boundaries in speech is highly uncertain). If these had been the only considerations affecting the transcription of words like wither, any of the three mentioned methods of transcribing the final part of the word would have served equally well. But there was the further consideration that derivatives formed by the addition of a vowel or vowel-initial suffix to many such words have pronunciation variants one of which has one syllable less than the other, withering and withery, for example, being either three-syllabled or two-syllabled. Such variation occurs often, especially in the present participle of verbs. By treating the sound of the er in the three-syllabled pronunciation of withery as a syllabic consonant \r\ we can show the variation thus: \with -r(-)ē\, which means \ˈwith -r-ē, ˈwith -rē\ or \ˈwith -rē, ˈwith -r-ē\, the parentheses indicating the alternative presence or absence of the phonetic item symbolized between them. For the present participle withering, following after wither transcribed \ˈwith -r\ at the beginning of the main entry, the indication of this variation can even be reduced to \-r(-)ing\. If the transcription of wither had been given as \ˈwith -ər\, a parentheses formula for the two variants of withering would have entailed either \ˈwith -(-ə)r(-)ing\ or \ˈwith -(ə-)ring\, the first clumsy with its two pairs of parentheses one of which encloses two characters instead of one, the second having two characters parenthesized instead of one and having one division changed from what it is in the form without a suffix. The alternative to a parentheses formula would have been one of the two fairly long ones \ˈwith -ər-ing, ˈwith -ring\ or \-ər-ing, ˈwith -ring\. Omission of the first syllable from the second variant in the last formula would have been ambiguous and even if done would have yielded a formula that is still longer than the \-r(-)ing\ formula. Primarily because the \-r(-)ing\ formula for the present-participle variation is the shortest one we were able to contrive for use in a transcription system in which syllable boundaries are indicated, \r\ is treated as being capable of being a syllabic consonant in the school dictionaries, in one of which the formula is extensively used.
The present book shows for well-known entries the pronunciation of both r ‑keepers and r ‑droppers and does not attempt to show boundaries for all syllables. Because of these two practices in this book we transcribe wither as \ˈwith -ə(r)\ and withering as \ˈwith (ə)riŋ\. This last is to be regarded as merely one of several possible ways of representing exactly the same two variant pronunciations that are represented by \ˈwith -r(-)ing\ in our other dictionaries. We could achieve somewhat closer parallelism between the two systems by representing in this book the r ‑keepers' pronunciation of wither as \ˈwith ə r\ and the two variants of withering as \ˈwith (ə )riŋ\ (paralleling \ˈbət(ə )niŋ\ for buttoning ), but a transcription \ˈwith ə (r)\ is impossible since \ˈwith ə \ is impossible.
In this dictionary we use the same symbols for the r ‑keepers' pronunciation of the stressed ur in the first syllable of murmur and the unstressed ur in the second syllable of that word, \ˈmərmər\. In the school dictionaries we do not, transcribing \ˈmərm-r\, for two reasons:
(1) Primarily, there might be strong opposition to showing a syllabic consonant in a stressed syllable, since the average dictionary user has never encountered a system that so employs a syllabic consonant.
(2) Although the sound of the stressed ur in murmur is a single sound, the sound of the ir in firm and the ur of fur is often a double sound when these monosyllables are in prepausal position, particularly if they are emphasized, and transcribing the ir of firm, the ur of fur, and the first ur of murmur as two sounds has phonetic justification. The vowel of ray is apt to be diphthongal and the vowel of the first syllable of raking is apt to be monophthongal in most American speech, but sound phonetic practice calls for regarding these variations as nonsignificant and using the same symbol or symbols in both situations.
We transcribe the ir of firm and firmer as if it were pronounced as two successive sounds. We are not certain that this is the only phonetic item answering to a vowel letter and an r in the spelling that is sometimes pronounced as a single sound. We often pronounce the ar of army so, and if we use a single symbol for the ir in firm (er) we ought also to use a single symbol for the ar of army. But to add symbols to a pronunciation alphabet unnecessarily seems unwise and useless. It is simpler and adequate to use a vowel symbol followed by \r\, with the understanding that the \r\ may represent either a second, independent sound or merely a modification of the preceding sound. Of the four consonants — \l\, \n\, \m\, \ng\ — that we treat as capable of being syllabic, the first three are shown as syllabic in the school dictionaries in many environments in which they are treated as \ə\ plus consonant in this book. The chief reason for this is the same as for treating the er of wither as a syllabic \r\: if we show the pronunciation of hyphen as \ˈhīf-n\ and of rival as \ˈrīv-l\, we can show the two variant pronunciations of the present participles by simply parenthesizing a single character in the brief formulas \-n(-)ing\ and \-l(-)ing\. When syllable divisions in transcriptions are dispensed with, as in this book, it is possible to treat the al of rival as vowel plus \l\, the double pronunciation of the present participle being shown by the simple parenthesizing of a single character \ˈrīv(ə)liŋ\.
If the reader will pronounce each of the following items with the consonant between the two vertical bars as a syllabic consonant and then as \ə\ plus consonant, there should be no doubt that the two pronunciations of each word are markedly different:
(1) \ˈkit|n|\ (kitten )
(2) \ˈred|n|\ (redden )
(3) \ˈkēp|m|\ (keep 'em )
(4) \ˈgrab|m|\ (grab 'em )
(5) \ˌkəp|m|ˈsȯsə(r)\ (cup and saucer )
(6) \ˈseb|m|\ (seven )
(7) \ˌläk|ŋ|ˈkē\ (lock and key )
(8) \ˌeg|ŋ|ˈkəp\ (egg and cup )
For items (1) and (2) a pronunciation \ən\ instead of \ə n\ is heard with some frequency especially in formal speech, but even in such speech it does not sound quite natural to many listeners. For items (3) and (4) both \ə m\ and \əm\ sound natural. For items (5) and (6) the \əm\ pronunciation sounds utterly unnatural, as does the \əŋ\ pronunciation for items (7) and (8). For the phonetic entity between vertical bars in items (5)–(8) \ən\ is a perfectly natural alternative (in item (6), when \v\ precedes) and if any experimenters believe that \əm\ and \əŋ\ sound normal in these items they are probably pronouncing or hearing \ən\ instead. In all eight items the first vertical bar is immediately preceded by a stop consonant that is homorganic with (i.e., that has the same basic articulation as) the consonant between the two. This book treats the nasal consonants \n\, \m\, and \ŋ\ as capable of being syllabic after such a homorganic consonant. If the preceding homorganic consonant is in turn preceded by a consonant, in many cases the first of the three consonants inhibits the syllabicity of the third. Thus London is nearly always \ˈləndən\, although fountain is as often \ˈfau̇ntə n\ as \ˈfau̇ntə̇n\.
The same criteria used for the possible syllabicity of the nasal consonants after homorganic stops point to the advisability of treating \l\ as capable of syllabicity after the homorganic stops \t\ and \d\ and also after the homorganic nasal \n\ (which is classifiable as a nasal stop), as in metal, meddle, and final. As with \n\, pronunciation of these words with \əl\ instead of \ə l\ is sometimes heard in formal speech. With most American speakers such pairs as metal : medal are quite identical in pronunciation, the consonant preceding the \l\ being a variety of \d\ (see the section on \d·\ ).
We have found it difficult to decide about the transcription of the group of words in which the consonant sound preceding \n\, \l\, or \m\ is a homorganic consonant other than a stop. 1
The following words illustrate this classification:
(1) \ˈbās|n|\ (basin )
(2) \ˈrēz|n|\ (reason )
(3) \ˈōsh|n|\ (ocean )
(4) \ˈvizh|n|\ (vision )
(5) \ˈskəch|n|\ (scutcheon )
(6) \ˈpij |n|\ (pigeon )
(7) \ˈnāth|n|\ (Nathan )
(8) \ˈhēth |n|\ (heathen )
(9) \ˈgris|l|\ (gristle )
(10) \ˈēz|l|\ (easel )
(11) \ˈsōsh|l|\ (social )
(12) \ˈyüzh|l|\ (usual )
(13) \ˈsach|l|\ (satchel )
(14) \ˈaj|l|\ (agile )
(15) \ˈeth|l|\ (ethyl )
(16) \bə̇ˈkwēth |l|\ (bequeathal )
(17) \ˈstəf|m|\ (stuff 'em )
(18) \ˌdef|m|ash|n|\ (defamation )
(19) \ˈlēv|m|\ (leave 'em )
(20) \ˈsev|m|\ (seven )
The two chief kinds of data used in determining what our transcription of each of these types of word should be were (1) our records, accumulated over an extended period of listening, of our impression as to whether a consonant was syllabic or was preceded by a vowel \ə\ and (2) the impression arrived at after much pronouncing by ourselves such items (often containing made-up words or nonsensical names) as May Sinnott: mace a knot: Maysa Nott: Mason Ott, which could differ from one another only in the presence or absence of a vowel before the consonant being explored (items containing a pause were out of order). After repeated alternate attempts to pronounce \ən\ and then \ə n\ in the consonant successions in items (3)–(6), the successions all sound alike whether in our made-up test items there is a word boundary before the first member of the consonantal succession, after the second member, or between the two. Such an outcome determines an arbitrary choice of either \ən\ or \ə n\ for these words, and the equivalent of \ə n\ is used in the school dictionaries; in this book the syllable core of these words is represented, except for a few interjections, in all cases where the consonants \n\, \l\, \m\, and \ŋ\ are not involved, by a vowel symbol. Such representation also avoids transcribing the last part of fission as \-ə n\ and the first part of another as \əˈn-\, with their suggestion that fission others and fish another's do not sound alike.
Words of the type illustrated in (1) and (2) are transcribed with ə n because pairs like Mason Ott and May Sinnott seem distinct and because our records contain notations of pronunciations of words like reason with overpronunciations having \ən\ at the end. The difference between \zə n\ and \zən\, however, is by no means as easy to hear as the difference between \də n\ and \dən\, and when an unstressed vowel precedes the \s\ or \z\ a clear distinction between \sən\ and \sə n\ seems impossible, Jefferson Ott and Jeffersa Nott and Jeffer Sinnott all sounding the same to us. Hence we transcribe \ˈgärsə n\ (Garson ) but \ˈjefərsən\ (Jefferson ).
Items (7) and (8) are more difficult, but a distinction can be heard between \thən\ and \thə n\ and between \th ən\ and \th ə n\; \ən\ is regular when a vowel follows the \n\ without pause, being heard in unearth a nape, unearth an ape, Bertha Nape, and earthen ape; and either \ən\ or \ə n\ is apt to occur when a consonant or a pause follows the \n\, \ən\ and \ə n\ in such position being in free variation. We have therefore shown \ən\ after \th\ and \th \, to the exclusion of \ə n\.
To make \əl\ and \ə l\ in items (9)–(16) sound distinct and like normal English is also difficult, and we accordingly transcribe \əl\ after the consonants listed there. A number of transcribers show a syllabic \l\ in words like gristle and dazzle, but this may be primarily because of the analogy of the orthography, in which there is no vowel letter immediately before the l.
After \v\ in items (19) and (20) \ə m\ and \əm\ are possible and are quite distinct: \ˈsevəm\ for seven would sound strange to probably all educated persons whose native language is English whereas \ˈsevə m\ sounds quite natural to nearly all and is a part-time pronunciation of many, most of whom are entirely unaware that they are hearing or pronouncing an \m\ at the end of the word, are unwilling to believe that any educated speaker so pronounces, and actually denounce the pronunciation as worthy only of the most illiterate.
The syllabic \m\ of \və m\ has an articulation different from that of most \m\ˈs in English, being formed not with the lips but with the lower lip and the upper front teeth, by assimilation to the \v\. The same kind of \m\ is heard in nymph and symphony. Its lip-teeth articulation is an almost inevitable result of its environment and in most transcription requires no special symbol. \və m\, however, has a very strong tendency to become \bə m\ which is somewhat easier to utter and has the usual two-lip variety of \m\.
Certain common speech items having v and m in the orthography may have any of the three variants \vəm\, \və m\, \bə m\, as government, leave 'em (these three pronunciations for government are not limited to r ‑droppers). Possibly even an uncommon item like Novum Organum would be heard with all three. It seems likely that an item like Neva Martin, in which what precedes the letter M is a word that in isolation is pronounced with a final vowel, would usually have only the variant \vəm\, although it might be generalizing too much to rule out \və m\ and \bə m\ altogether. For all v … m items a pronunciation \bəm\ does not sound natural.
A number of v … n items have any of the three variants \vən\, \və m\, \bə m\, as seven, eleven, davenport. For these, either \vəm\ or \bəm\ does not sound natural. b … m items like rob 'em are either \bə m\ or \bəm\. For these, \və m\ or \vəm\ would not sound natural.
Items of the type illustrated in (17) and (18) — stuff 'em, defamation — do not offer as many complications as do those just discussed. The \f\ does not tend to become \p\, as \v\ does to become \b\, and the \m\ does not tend, to an extent at least that would appear to require notice in transcriptions in this dictionary, to appear in words in which the corresponding orthograpic item is n, as hyphen. The variants \fəm\ and \fə m\ both occur freely for items like stuff 'em and sophomore (the last also has a two-syllable variant) but \fə m\ would probably be a pretty rare variant in an item like Josepha Moore.
In words exemplified in the following list, there is a syllable core between an \l\, \n\, or \m\ and the next preceding consonant, which is not homorganic (successions in which a nasal is preceded by a nasal are discussed later; no words in which unstressed \əŋ\ is a possibility are shown because such a succession probably does not occur in normal English, and \ə ŋ\ occurs only as an occasional variant):
(1) triple
(2) babble
(3) fickle
(4) struggle
(5) baffle
(6) bevel
(7) camel
(8) happen
(9) ribbon
(10) weaken
(11) wagon
(12) stiffen
(13) seven
(14) bottom
(15) madam
(16) oakum
(17) ligament
(18) blossom
(19) chasm
(20) Gotham
(21) rhythm
(22) Beecham
(23) Bridgham
(24) Gresham
Although words of the type triple and babble, and of the type chasm and rhythm, have been shown as having syllabic \l\ or syllabic \m\ in many books containing transcriptions, an \əl\, \ən\, or \əm\ in these words distinct from an \ə l\, \ə n\, or \ə m\ is undetectable, and we accordingly transcribe all these words with \əl\, \ən\, \əm\. Some of these words having an n in the spelling have pronunciation variants in which, by assimilation to the preceding consonant, the \n\ becomes \m\ (homorganic with preceding \v\, \p\, or \b\) or \ŋ\ (homorganic with preceding \k\ or \g\). In such variants the \m\ or \ŋ\ is syllabic and is transcribed \ə m\ or \ə ŋ\. Such variants are shown for common words (as happen and seven ), but for other words they are to be understood as occasional pronunciations largely limited to extremely rapid speech or to predisposing environments. (Thus, \-gə ŋ\ would be less frequent for dragon immediately before a pause than for dragon in dragon claw, an environment in which the nasal is not merely preceded but also followed by a consonant having the same place of articulation as \ŋ\).
It seems likely that some who transcribe words like triple, fickle, chasm, and prism with a syllabic consonant \l\ or \m\ do so in continuance of a tradition that originated in a day when transcribers stood more in awe of spelling than the more enlightened do today. Transcriptions of rifle and supple that do not parallel those of rival, opal, and oval, and transcriptions of chasm and prism that do not parallel those of blossom and gruesome are suspect. There are of course passages of poetry in which the pronunciation of an ‑sm word as \-zəm\ would roughen the meter — a phrase like “a chasm of granite rock” in a line requiring iambics. But what is required in such a passage is not the use of \ə m\ instead of \əm\ but the use of a pronunciation in which the \z\ and the \m\ stand to each other in the same relationship as in the word jazzman. Such a pronunciation is easy enough to make clear by the use of spaces in the transcription: \ə kaz məv granə̇t räk\. Any attempt at the entry chasm to transcribe such a pronunciation of the sm would probably elude even most of the phonetically sophisticated, unless an explanation were given. Some might say that such a pronunciation is merely an example of poetic license and as deserving of ready dismissal as the pronunciation \th \ required for the definite article in “and dashed th' ambitious hopes” or the pronunciation required for loathed in “the loathèd hut they left behind”. Such elision of a vowel between certain consonants and a following word-final \m\ that precedes an unstressed vowel occurs in prose utterance as well (the more rapid the utterance, the more likely its occurrence) and it also occurs before word-final \l\, \n\, and even \ŋ\, despite the fact that its occurrence before \ŋ\ would require that \ŋ\ be regarded as preceding a vowel in the same syllable, something that \ŋ\ is supposed never to do in normal English. Thus such pronunciations as the following are heard frequently:
\ə mit nə(v) wu̇l\ (a mitten of wool )
\tə sēz nə rōst\ (to season a roast )
\ə kak lə(v) gēs\ (a cackle of geese )
\th ə pē plə(v) rōm\ (the people of Rome )
\thə {sev nə(v) / sev mə(v) / seb mə(v)} kləbz\ (the seven of clubs )
\th ē ə̇nthüzēaz mə vȯl\ (the enthusiasm of all )
\th ə rith mə velēəts līn\ (the rhythm of Eliot's line )
\ə̇t wōn(t) hap {nəgen / məgen}\ (it won't happen again )
\ə wag ŋə dā\ (a wagon a day )
\bak ŋəp th ə kär\ (backing up the car )
Except for an occasional entry word (as ²couple ) that has prevocalic occurrence in a frequently pronounced phrase, such variants are not shown. Such information is more likely to be sought in a discussion of phonetics than at individual entries in a dictionary, and constant repetition of a formula adequate to make this difficult-to-transcribe variation clear is felt to be unjustified.
The preceding remarks have touched on the pronunciation of a word like chasm in prevocalic position. Also to be found in poetry are such passages as “the chasm 'twixt rock and rock” (Browning), with chasm preconsonantal and in a position where optimally smooth meter requires a monosyllable. Uttering chasm here in a way that would be normal in any prose passage produces a very jarring rhythm.
Transcriptions of chasm that have appeared in some of our earlier dictionaries are (kăzm) and (kăz'm). (kăz'm) parallels the transcription of words like dazzle and button in having an apostrophe (symbol indicating a following consonant as syllabic) before the final consonant but differs from them in having no stress mark before the apostrophe (in our earlier dictionaries a stress mark following any syllable except the last indicated not merely stress but also a syllable boundary). The transcribers of both (kăzm) and (kăz'm) may have been influenced by an awareness of the monosyllabicity desirable for chasm in some poetic passages and may have intended these transcriptions to represent either a disyllabic or a monosyllabic pronunciation. The orthography may have caused them not to realize that the variations occurring for chasm are paralleled in words like couple and mitten (which have a vowel letter either after or immediately before the last consonant whereas chasm has such a letter in neither position). The briefest possible transcription for such variation is \ˈkaz(ə)m\, \ˈkəp(ə)l\, \ˈmit(ə )n\, although the intent of such a transcription would probably be lost on most dictionary users.
Many or most r ‑droppers have syllabic consonants not only in words of the type cotton and curtain (in which the orthography has a vowel letter or vowel letters between the two consonants involved) but also in words of the type modern \-də n\, pattern \-tə n\, lantern \-ntə n\, international \¦intə n-\, utterly \ˈəd·ə lē\, and government \ˈgəvə mənt, ˈgəbə mənt\ (in which there is a vowel letter and an r between the two consonant letters in the orthography). The two pronunciations just shown for government occur frequently in the speech of r ‑keepers as well. Although it has been claimed or implied in transcriptional practice that there are r ‑droppers who consistently keep the members of such a pair as Patton : pattern distinct in their pronunciation, making the first \ˈpatə n\ and the second \ˈpad·ən\ or \ˈpatən\, there can be no doubt that pronunciations such as \ˈpatə n\ for pattern are common in r ‑dropping speech in both the U.S. and Great Britain.
Between some successions of two sounds, the minimum transition in English is a non-syllable-forming one: for example, the transition from \ī\ to \n\ in brine, from \r\ to \n\ in r ‑keepers' barn, from \g\ to \d\ in drugged, from \l\ to \n\ in one pronunciation of kiln, all monosyllabic words. Between other successions of two sounds (the successions discussed in some detail in the preceding paragraphs), the minimum transition is a syllable-forming one: for example, the transition from \d\ to \n\ in hidden, from second \p\ to \l\ in people, from \z\ to \m\ in chasm, from \k\ to \r\ in acre 2 . When both a minimum and a more-than-minimum transition can be heard, a vowel is present in the more-than-minimum transition and a vowel symbol should be shown in the transcription. Thus Bryan is \ˈbrīən\ not \ˈbrīə n\ (compare monosyllabic \ˈbrīn\ brine), baron is \ˈbarən\ not \ˈbarə n\ (compare monosyllabic \ˈbärn\ barn), rugged is \ˈrəgəd\ or \ˈrəgid\ not \ˈrəgə d\ (compare monosyllabic \ˈdrəgd\ drugged ). Successions incapable of less than disyllabic transition offer the complications that some are clearly capable of both a minimum disyllabic transition and a more-than-minimum disyllabic transition (\d\ and \n\ for instance, as in hidden ) whereas others are not. Some phoneticians, feeling it unnecessary to take cognizance of such transitional variation, transcribe alike all transitions that are disyllabic, either treating the second consonant of all such successions as having a vowel before it or as being syllabic. Transcribers who record both types of transition do not always agree to which classification a given succession belongs. Those who wish to distinguish them should use \ə \ between consonant symbols for the minimum transition and \ə\ for the more-than-minimum — for example, \ˈhidə n\ for the minimum, \ˈhidən\ for the more-than-minimum, in hidden.
For words like linen, allyl, minimum, and goaded, in which the two sounds of the succession are identical consonants, the transcription is with a vowel (\ə\ or in many cases either \ə\ or \i\) rather than \ə \, the transition being more-than-minimum. The minimum transition for such a consonant pair is that heard in a word like meanness, where the transition from \n\ to \n\ consists of a mere holding or prolongation of the articulation of the \n\ that is also regardable as a doubling of the consonant and is usually transcribed by writing the symbol twice. Citing meanness (a word with a vowel both preceding and following the consonant pair) violates the criteria announced earlier as a basis for this particular part of our discussion, but it is only in such words that this type of transition ordinarily occurs in English. In some languages, however, it occurs within the bounds of a single syllable as a regular phenomenon, and is occasionally used in English for effect. Thus Carl Sandburg pronounces Lincoln \ˈliŋŋkən\ in reading from his works.
Although there are those who use transcriptions of the type \ˈkümə n\ common and \ˈvenə m\ venom (just as there are those who use transcriptions of the type \ˈbrīə n\ Bryan and \ˈmelə n\ melon ), such transcriptions are out of line with the criteria we have used in that they employ the symbol of syllabicity for a more-than-minimum transition. The minimum transition for two different nasals is one in which the mouth closure for the second is simultaneous with a part or all of the closure for the first. Presumably only the articulation that is nearest the breath stream is acoustically identifiable, the outer one being blocked off by the inner one from forming any part of the resonance chamber. Such transitions do not occur in normal English nor in any of the languages with whose phonetics we have any firsthand acquaintance but they are quite easy to make. Mischa Elman once pronounced his surname over the radio as \ˈelm\ plus a nasal \n\ whose articulation began during the closure for \m\ and was held until the pause that follows the surname, as was also apparently the closure for the \m\. Whether this pronunciation should be considered disyllabic or monosyllabic is something easily argued.
\ə \
(following \l\, \m\, \r\; for ə preceding \l\, \n\, \m\, \ŋ\, see the sections \ə l\ , \ə n\ , \ə m\ , \ə ŋ\ ) in some foreign languages, of which French is the best known, two consonants the second of which is \l\, \m\, or \r\ can occur at the end of a word without any vowel between them. However, whereas in English the second member of the pair is voiced and the pair is disyllabic, in French the second member of the pair is voiceless (even when the first member is voiced) and the pair is monosyllabic. We indicate such a pronunciation of a pair of consonants by placing the symbol ə after rather than before the second member. Examples from French of words for which such a pronunciation is one variant:
couple \küplə \
table \tȧblə \
prisme \prēsmə , ‑ēzmə \
titre \tētrə \
mordre \mȯrdrə \
In French such words may have any of four pronunciation variants, representable in our system as follows:
table \tȧbl(ə ), ‑b(lə)\
The four variants shown are
1. \tȧblə \
2. \tȧbl\
3. \tȧblə\
4. \tȧb\
In French prose, 1 occurs before a pause, as does 4 chiefly in rapid or informal speech; 2 occurs before a word that begins with a vowel sound and that follows without pause; 3 and 4 occur before a word that begins with a consonant sound and that follows without pause, 4 occurring esp. in rapid or informal speech and in set phrases such as maître d'hôtel. The meaning of a transcription like 2 will be misunderstood by many, who will interpret it as disyllabic. In the following examples, spaces are used to indicate where the syllable boundaries are usually regarded as being in French, in order to facilitate an understanding of our transcriptions. Examples:
1. amour-propre \ȧ mür prȯprə —three syllables \
2. la table est belle \lȧ tȧ ble bel —four syllables \
3. table de bois \tȧ blə də bwä—four syllables \
4. amour-propre
\ȧ mür prȯp—three syllables\
table de bois
\tȧb də bwä—three syllables
\
\əi\
as in a pronunciation of bird that is widely used in greater New York City and in a strip of the deep South extending from North Carolina to Louisiana. This pronunciation is commonly represented by the letters oi, as in the spelling boid for bird, but the \ȯi\ suggested by this spelling is rarely or never the pronunciation actually used. In the South this pronunciation appears to be less regarded by educated speakers as one to be avoided than in the New York City area.
\əi\ does not occur as a variant of \ər\ and \ə̄\ in words having no sound after the \ər\ or \ə̄\, nor in formations based on such words — that is, \əi\ does not occur, in educated speech at least, in words like spur, spurred, spurring, furry. Nor does \əi\ occur in educated speech as a variant of \ər\ in words in which \ər\ is followed by a vowel, as hurry, flourish.
\ə l\
the usual sound of the le of battle and muddle, of the al of vital, of the el of model and funnel, of the ile of futile. The \ə \ is a diacritic and \ə l\ is a single sound, not two, an \l\ sound that serves as the nearest thing to a vowel in its syllable. Especially on formal occasions and with speakers not fully sure of themselves the two sounds \əl\ do sometimes occur instead of \ə l\. The \əl\ variant is usually not shown. When a word ending in \t\, \d\, or \n\ precedes without pause, words shown as beginning \əl-\ (as allow, alone ) may have \ə l-\ instead, as in not alone.
\ə m\
often heard for the words 'em and and \ when \p\, \b\, or \v\ precedes, as in stop 'em, grab 'em, save 'em, up and down, rub and buff, stove and poker. A preceding \v\ may be assimilated to \b\, as in the third and sixth examples above; if the \v\ remains \v\, the \ə m\ articulation is lip-teeth rather than lip-lip. \və m\ and \bə m\ are also often heard for the ven of words like seven and eleven and for the vernm of government. The \ə \ is a diacritic and \ə m\ is a single sound, not two, an \m\ sound that serves as the nearest thing to a vowel in its syllable. The two sounds \əm\ are not heard as a variant for \ə m\ when the orthographic correlate is n (government is not an exception because the orthographic correlate there is m not n, the first \n\ being lost possibly through the dissimilatory influence of the second \n\). When a word ending in \p\, \b\, or \m\ precedes without pause, words shown in the vocabulary as beginning \əm-\ sometimes have \ə m-\ instead, as in save ammonia.
\ə n\
as the usual an of Satan and en of wooden. The \ə \ is a diacritic and \ə n\ is a single sound, not two, an \n\ sound that serves as the nearest thing to a vowel in its syllable. Especially on formal occasions and with speakers not fully sure of themselves the two sounds \ən\ do sometimes occur instead of \ə n\. The \ən\ variant is usually not shown. When a word ending in \d\ or \t\ precedes without pause, words shown as beginning \ən-\, \ən-\, or \in\ (as announce, engage ) may have \ə n\ instead, as in made announcements, can't engage.
\ə ŋ\
sometimes heard as a variant for vowel plus \n\ or \ŋ\ when \k\ or \g\ precedes and somewhat more often when \k\ or \g\ also follows, as in chicken coop, breaking ground, dog and gun. The \ə \ is a diacritic and \ə ŋ\ is a single sound, not two, an \ŋ\ sound that serves as the nearest thing to a vowel in its syllable. When a word ending in \k\ or \g\ precedes without pause, words shown as beginning \ə̇nk-\ or \ə̇ng-\ sometimes have \ə ŋk\ or \ə ŋg\ instead, as in break engagements, big encouragement.
\ər\
(see also the section on \ər·, ə·r\ ) when in a syllable bearing either of the stresses ˈ ˌ and when word-final or followed by a consonant sound, \ər\ is the sound used by r ‑keepers for the ir of stir, the ear of search, or the ir of circularity. When the \ə\ of \ər\ is not preceded by a stress mark, \ər\ is the sound used by r ‑keepers for the er of walker and by both r ‑keepers and r ‑droppers for the ar of arise, the er of veneration, and the er of bakery when that word is pronounced in three syllables and not two.
Many transcribers regard what we transcribe as \ər\ as a single sound (stressed or unstressed) and use a single symbol instead of \r\ preceded by a vowel symbol. Such transcribers regard the articulation as a single one in which the body of the tongue is in the position for some such vowel as \ə\ or \ə̄\ at the same time that the tip is in substantially the position for the sound following the \n\ in unroll. Even if the pronunciation that we symbolize \ər\ is one sound, there appears to be no great objection to transcribing it as if the components of the articulation were successive rather than simultaneous, and such a transcription is now the usual practice of many linguists in this country. Such a practice makes it possible to show both the r ‑keepers' and the r ‑droppers' pronunciation of a word like winner by merely parenthesizing the \r\, thus: \ˈwinə(r)\. 3 Furthermore, it is by no means sure that, if \ər\ is a single sound, it is the only such sound in English: arc may be just as much a two-sound word as irk.
\ər·, ə·r\
used in words in which \ə\ is stressed and a vowel immediately follows the \r\. The first of the two transcriptions is also used at the end of words, since any word is capable of being followed by a vowel-initial word.
\ər·\ with a vowel sound following is the same sound as that written without the centered dot when a consonant sound follows — that is, it is the sound used by r ‑keepers for the ir of bird or the irre of stirred. It is also the sound used by r ‑keepers for the irr of stirring (most U.S. r ‑droppers have the same sound in this word, a smaller number have \ə̄r\) and by r ‑keepers for the ir of Sir Albert (some U.S. r ‑droppers have the same sound, some say \sə̄ˈalbət\, and fewer say \sə̄ˈralbət\, the usual pronunciation in southern British speech, as is \ə̄r\ for stirring ). \ər·\ is also the pronunciation used by most r ‑keepers for the urr of hurry and the our of courage, words that differ from the linguistic items stirring and Sir Albert in not being composed of two meaningful English units.
Most U.S. r ‑droppers, speakers of southern British, and r ‑keepers in some areas of the U.S. pronounce the urr of hurry and the our of courage in a way that we transcribe \ər·\. In \ər·\ whatever we represent by the \ə\and the \r\ is in the same syllable: the syllable division occurs after the \r\. But in the pronunciation that we represent \ə·r\ the \ə\and the \r\ seem very definitely to be distributed between successive syllables. Therefore, we have made the distinction between the two by a difference in the placement of the centered period, a syllable-boundary indicator. One of the most expected places for the occurrence of the \ə·r\ pronunciation is in the circus barker's spiel “Hurry! Hurry! Hurry!” where, it seems likely, \ə·r\ is used even by speakers who ordinarily have \ər·\ in such words. This sound may be produced by saying the first two sounds of hum without sounding the \m\ and then adding the first two sounds of react.
In transcriptions of uncommon words rhyming with stir, like claqueur \klaˈkər(·)\, the parenthesized period means that the period is to be omitted when a consonant or a pause follows (as in \klaˈkərz\ claqueurs ) but is to be retained, without the parentheses, when a vowel sound follows without pause (as in \əklaˈkər·ə̇nth əˈbalkənē\ a claqueur in the balcony ). \-ˈkə·rə̇n-\ for this last phrase would be a pronunciation that does not ordinarily occur in educated standard speech.
In words as common as hurry both the \ər·\ and the \ə·r\ variant are shown, in the expectation that it will be chiefly in the more frequent words that phonetic detail will be sought in this book. But for a word having as little currency among nonspecialists as gurry the transcription is shown simply as \ˈgərē\, since either of two likely interpretations placed on the \ər\ will be acceptable.
In words like stirring, burry, formed by adding an active vowel or vowel-initial suffix to words of the type stir, the \ə·r\ variant is not usually heard, although it appears to be an occasional, and the usual British, variant of the noun furrier and a rare variant of furry (and therefore of the comparative adjective furrier ) in British speech. In words of the type occurrence, which are from, or which parallel words existing in, a foreign language, notably Latin, and which if formed in English were formed by using a suffix now relatively inactive in English, an \ə·r\ variant is quite regular with speakers who have \ə·r\ in hurry.
\ əu̇\
as in one pronunciation of bulb — see section on \l\ .
\əw\
this succession of symbols followed by a vowel symbol (as in \ˈstrenyəwəs\ for strenuous, \ˌsichˈwāshən\ for situation ) occurs instead of the \u̇\ or \ō\ followed by a vowel that is employed by a number of transcribers. The starting point in the reasoning that led to a decision to use \əw\ was a very strong conviction that the members, or that the shorter member and a part of the longer member, of such pairs as the following are exact rhymes (some of the words and names below are made up and some of the combinations are nonsensical):
silhouette
Scylla wet
Genoese
Jenna Weeze
Manuella
Manya Weller
Halloween
Valhalla wean
willowy Ona
Willa Weona
habituate
bitch a wait
Venezuela
Venn is a weal (or wale )
Granting the homophony of the members of those pairs, without any attempt to explore the phonetics of the matter, we transcribe \əw\ for the following two reasons: (1) Transcribers have been unanimous, or nearly so, in using \ə\ and \w\ in transcriptions of items of the sort we have shown in each pair as the second member, but the same transcriber ordinarily uses \u̇\ when the spelling has u and \ō \ when the spelling has o. If homophony is granted, then it must be that transcribers are using two symbols (\u̇\ and \ō\) for what is really the same sound or succession of sounds. Such variation violates one of the most fundamental tenets of transcription. (2) If we decide to transcribe silhouette and Scylla wet both \silu̇et\ or \silōet\ rather than \siləwet\, then providing variant transcriptions at both words of the type Scylla and words of the type wet that will take care of their pronunciation in isolation and their pronunciation in succession, with the w ‑initial word second, is a task so formidable as to be prohibitive. If we transcribe both silhouette and Scylla wet with \əw\, however, then the transcription for Scylla and for wet in isolation is also the transcription for them in succession in that order. Thus only at words of the type follow and value must variants be provided for isolation and succession differences where a variant in \-əw\ is provided to take care of the usual or a common pronunciation in items like following, follow up, valuer, value alteration.
\a\
as in hat, gap, has-been, have-not; some speakers have \a\ also in one, two, or all of the three words hash, hang, ask. Other speakers have in hash and ask a vowel that we symbolize \aa\ or \aaə\; they have the latter vowel also in halve and in the emphatic pronunciation of the verb can, words which may be distinct from the emphatic form of have and from the noun can (“container”), the latter two having \a\. Other speakers have in hash, hang, ask a vowel that we symbolize \ai\; they also have the latter vowel in halve and, if the word is in their vocabulary, probably also in salvy, words which if so pronounced are distinct in the vocabulary of most from the emphatic form of have and from savvy, the latter two usually having \a\. Still other speakers have in ask a vowel that we symbolize \ȧ\; they also usually have the latter vowel in one or more of the words halve, heart, salvy, words which then are distinct from the emphatic form of have, from hat, and from savvy, the latter three having \a\. See the sections on \aa(ə)\, \ai\, and \ȧ\.
The sound in the vowel of hat may vary according to the dialect, but our symbol for all varieties is \a\. To speakers who have \aa(ə)\ in their vocabulary the vowel used in hat in other dialects may sound to them like or much like their \aa(ə)\. The \a\ of speakers whose vocabulary includes both \a\ and \aa(ə)\ may sound to speakers of other dialects like or much like \ȧ\. The vowel of hat in Scottish dialect is a sound that has a quality like that often heard for the ear of heart in eastern New England or southern British speech or for the o of hot in most U.S. speech (that is, that has an \ȧ\ or \ä\ quality) but that is always very short. For the Scottish vowel, a symbol different from that for the standard vowel of hat is often employed in dictionaries. However, since the standard-English vowel of hat does not occur in the type of Scottish transcribed in this dictionary and it and the Scottish vowel of hat are not in contrast, there is no necessity for using a different symbol to transcribe the Scottish sound. The speaker of Scottish will use the sound natural to him in hat when he sees \a\ in a transcription, just as the American and the southern Englishman each interprets \ō\ for the vowel of toe in his own way, although their vowels are usually markedly different.
\ā\
as the vowel of day, fade, date, aorta, when emphatic usually more diphthongal in British speech than in most American speech. When it is a diphthong, the second element is \i\ and the first element is either the vowel of debt or the vowel used in day when that vowel is a monophthong.
\ä\
as used in cot and cod by those who pronounce these words differently from caught and cawed. The quality of the vowel of cot and cod varies, in length and in quality. In southern British speech, it is consistently short but in the U.S. may be quite long, especially before a voiced consonant. The vowel of most southern Englishmen and of many eastern New Englanders is pronounced with the tongue higher and further back and with the lips more rounded than the vowel of the average American, to whom the others' vowel may sound more like that of his caught or cawed. Our symbol \ä\ is to be interpreted as covering all varieties of vowel used in the English-speaking world for the vowel of cot and cod by speakers who pronounce these words differently from caught and cawed. Although this vowel is more often spelled with o in English than with any other letter, we use \ä\ rather than \o\ or an o ‑based character in our transcription for this reason: We must be able to show that in the speech of most or many Americans the first two italicized items in each of the groups of three below are identical in sound and that the third item in each group consists of the same sounds in the same order but with the additional sound \r\ present immediately after the vowel under discussion:
Fother gill
father
farther
bomb
balm
barm
bock beer
Bach
bark
mock
mach
mark
The first two orthographies in each group show that either \a\ or \o\ with or without a diacritic would be a symbol with precedent in English orthography for the words without r in the spelling or \r\ in the pronunciation. For this last class of words, however, \a\ is a better symbol than \o\ because \o\ before an \r\ suggests the vowel of cord rather than of card.
When one transcribes, as sound transcriptional practice requires, on the principle of one symbol to a sound, it is impossible to have all transcriptions of a language as orthographically lawless as English look right. A vowel with an articulatory position as far forward as that of fair and fare is usually represented orthographically by a or a letter group containing a and has traditionally been represented in transcriptions by \a\ with a diacritic over it. On the other hand, a vowel with an articulatory position as far back as that of all and saw is often represented orthographically by a or a letter group containing a and is represented in some pronunciation systems by \a\ with a diacritic over it. The use of \a\ with a diacritic over it to represent a vowel with an in-between articulatory position should therefore occasion no surprise. In a phonetic alphabet widely used by American linguists today, the vowel of cot is \a\. In the alphabet of the International Phonetic Association, which is used by many transcribers of American speech, the symbol used for the vowel in this word is one that resembles a script or italic a or Greek alpha.
For any who may require in their transcribing a symbol for the unrounded vowel that occurs in cot and a symbol for the rounded vowel that occurs in the same word and is distinct from the vowel of caught, we suggest \ä\ for the unrounded vowel and \o\ for the rounded. It was common practice in dictionaries, and is still the practice in some dictionaries, to use symbols for four distinct vowels in the range from front \ȧ\ to back \ȯ\ — symbols equivalent to \ȧ\, \ä\, \o\, and \ȯ\ respectively. Although there may be dialects of English in which a distinction is consistently made between four classes of words on the basis of four vowels representable by \ȧ\, \ä\, \o\, and \ȯ\, we have no reason to believe that there are speakers enough who make such a distinction to justify its notice in our dictionary. Perhaps a knowledge that the usual or a frequent British vowel and the usual American vowel in cot are different accounts for some American transcribers' employing four symbols rather than three. The desideratum in the formulation of a phonetic alphabet, however, is not the provision of a one-character symbol for every minute shade of sound but the provision of the barest minimum necessary to represent the distinctions consistently made in the speech of any one speaker. For distinctions any finer than this, the phonetician may use modifiers. When the explanation is made that \ä\ is the vowel of cot, whatever the quality of that vowel may be; that any whose vowel (in cart, for example, in the speech of r ‑droppers) is consistently between that of cot and the further-forward and higher vowel of cat have a vowel that we symbolize \ȧ\; that any whose vowel (in caught or quart, for example) is consistently between that of cot and the further-back and higher vowel of coat have a vowel that we symbolize \ȯ\, we have provided symbols enough to transcribe all the vowel distinctions that most speakers make in this range.
Of the best-known dialects of English, none appears to have more than three vowels in this range, and some appear to have only two. Any who may have concluded from the presence in southern British of a rounded vowel in cot which is distinct from the unrounded vowel of cart and from the even more rounded vowel of caught that there are four vowels between \a\ and \ō\ should observe that Daniel Jones and Henry Wyld transcribe only three. The number of such words that are phonetically distinct in southern British probably does not differ materially from the number that are distinct in the speech of U.S. r ‑keepers who do not level pairs like cot and caught. Such U.S. speakers make no distinction between such words as Spaak and Spock but southern British speakers make no distinction between such words as Spaak and spark. It seems likely that the mistaking of one word for another in this group is greater than average. If so, an explanation is to be sought not merely in the fact that some speakers have two vowels instead of the three of other speakers but also in the fact that with speakers who have the same number of vowels the distribution of the vowels is subject to much variation, as was just illustrated in the words Spaak, Spock, and spark. Ordinary words known in all dialects cause little difficulty, for the same reason that such identically pronounced words as meet and meat seldom cause difficulty: the context in nearly every case leaves the hearer in no doubt as to what word the speaker is using. But unknown or little-known words — surnames and the names of small geographical entities, for example — can sometimes cause perplexity.
In our transcriptions we have attempted to use only the minimum number of symbols required to show the distinctions that occur in any one major dialect of English. Since the discussion that follows involves the comparative phonetics of a group of words showing a high degree of variation and overlapping in some of the major dialects of English, we will use symbolizations that do not occur in the dictionary. Those that do we will enclose in lightface square brackets rather than in our usual reverse slant lines. [ȧ] represents a vowel that has a tongue position somewhat lower and further back than \a\; it is the vowel usually heard in part and usually or often heard in pass in eastern New England speech. [ä] is the vowel that the majority of Americans use in pot. [o] is the vowel that is used in southern British speech in pot. It differs from the usual American vowel in the same word in being pronounced with the lips rounded and often or usually in having a tongue position slightly higher and further back. [ȯ] is the vowel used in the first syllable of order and in sawed by speakers with whom sawed and sod are distinct. Doubling a vowel symbol indicates a lengthening of the sound it represents; [oo] indicates a vowel of the same quality as [o] but longer; [oo] does not represent the vowel of hoot or hook.
Such pairs as order ¦ ardor, born ¦ barn, in which a consonant sound immediately follows whatever pronunciation is given the or or ar, are no more than sporadically leveled in any well-known dialect. There appears to be some leveling in the Southern Mountain dialect and perhaps in Utah. In a section of southern Illinois the first part of the word Borneo is pronounced exactly like barn. Speakers of the dialects of the Atlantic and Gulf coasts and of Texas, from New York City to El Paso, in which [o] occurs with some frequency in the first syllable of ardor and in barn, with or without an \r\ following the vowel, are apt to be misunderstood as pronouncing order and born instead by the speakers of dialects in which ardor and barn have [ä] or [ȧ]; but the percentage of misunderstandings appears to be no higher when both speaker and listener have the same distinction than in the case of other vowel pairs equally close in articulation.
With the majority of American speakers, such pairs as Mahler : Moller, father : Fother gill, Brahms : Brom's, khaki : cocky (when the a of khaki is not \a\), and Bach's : box (when the ch of Bach's is not \ḵ\) are exact rhymes. When they are, the vowel of both is usually [ä], but in Southern Mountain speech may be [o]. These two vowels are represented by \ä\ in our dictionary transcriptions.
In some dialects of English (especially those in which r ‑dropping occurs, as eastern New England, southern Coastal, southern British, and New York City), the vowels of such pairs are usually or often not homophonous. The stressed vowel of Mahler may be [ȧ] (as in eastern New England and southern British), a vowel articulated further forward in the mouth than the stressed vowel of Moller, lacking the rounding that usually or often accompanies the vowel of Moller, and in southern British at least usually longer than that vowel. The vowel of Mahler may be [ää] or [äə] (as in southern Coastal and New York City), a vowel having the same quality at the beginning as the vowel of Moller but appreciably longer because of prolongation or because of the presence of a slight [ə] off-glide at the end. The vowel of Mahler may be [oo] (as also in southern Coastal and New York City), a vowel with the quality of [o], the rounded vowel heard in southern British not, but longer. All of these three vowels or vowel characteristics that make the stressed vowel of Mahler distinct from the stressed vowel of Moller are symbolized in the dictionary pronunciations by \ȧ\. If only southern Coastal and New York City speech were being transcribed, the vowel of Mahler, when it differs from that of Moller, could very well be transcribed as \ää\. This could be made to provide for the [oo] variant as well by an explanation that lengthened \ä\ is often accompanied by rounding. That Mahler and Moller are sometimes homophonous and sometimes not could be shown by the transcriptions \ˈmä(ä)lə(r)\ and \ˈmälə(r)\ respectively. Speakers of dialects in which the stressed vowel of Mahler is [oo] may easily be misunderstood by hearers not familiar with the dialect as saying mauler instead of Mahler. Whether within such dialects the stressed vowels of Mahler, mauler, and Morley are ever completely leveled is doubtful. With southern speakers who have [oo] in Mahler, the stressed vowel of mauler (but not, it appears, that of Morley ) is often a diphthong [ȯu̇], acoustically quite distinct from [oo].
In r ‑keeping speech pairs like Spaak : spark are distinct because one member contains an \r\ and the other does not. In r ‑dropping speech such pairs usually are homophones, but sometimes in such speech Spaak goes with Spock rather than with spark, as it does in most r ‑keeping speech. Our symbol for the vowel of Spaak when it is homophonous with that of Spock is \ä\; for the vowel of Spaak when it is not homophonous with that of Spock, \ȧ\.
In r ‑keeping speech pairs like spark : Spock, Farley : folly, harmony : hominy are distinct because one member contains an \r\ and the other does not. In some such dialects (as Middle Atlantic and the r ‑keeping speech of Texas), the vowel of spark is apt to be rounded ([ˈspork]) whereas the vowel of Spock is apt to be unrounded ([ˈspäk)]), but we can transcribe \ˈspärk\ and \ˈspäk\ and explain that \ä\ followed by \r\ may be rounded. Such an explanation holds both for words like spark, in which a consonant follows the \r\, and for words like moral and horrid, in which a vowel follows the \r\.
In some r ‑dropping dialects (as eastern New England and southern British) a pretty consistent qualitative distinction is made between the vowels of spark and Spock, and often in addition a quantitative or durational distinction as well. If the quantitative difference is lacking, the qualitative one sets the words apart. (Incidentally, the eastern New England vowel of spark is usually markedly different in quality from the southern British vowel in the same word, \ȧ\ standing in the former for a vowel much closer to that of cat and \ȧ\ being in the latter a vowel much the same in quality, and often in quantity also, as the vowel that most Americans have in cot. ) Although some American phoneticians have observed that the southern British vowel in cot doesnot always have the rounding that the writings and transcriptions of British phoneticians seem to claim for it, the extent of the leveling of spark and Spock is beyond much question not as great as in the dialects next discussed.
With r ‑dropping speakers (as in the coastal Southern states and in New York City) who never have more than a quantitative difference between the vowels of spark and Spock, this difference of duration frequently breaks down, with the resultant homophony of such pairs. Some who have written on this type of leveling attribute it to a lengthening of the [ä] of the r ‑less member of the pair, chiefly before voiced consonants, the tendency to lengthening being greater before these than before voiceless consonants. But the leveling may be due also to a shortening of the vowel of spark. The tendency of a vowel to be short in duration increases with its distance from the last syllable of a word, and such words at least as harmony and hominy, in which the [ä] is in the third syllable from the end, both sometimes have a fairly short [ä] and are homophonous. Such leveling is not indicated in our transcription. A variant \ä\ is to be understood as occurring with varying frequency for all words transcribed with the variants \är\ and \ȧ\ (in uncommon words, with \är\ only) before a consonant. Our omission of the third variant does not mean that we regard the variant as substandard but rather reflects the uncertainty that exists as to the extent of the variation.
In some areas (as western Pennsylvania, eastern New England, and parts of the Far West) the members of such pairs as cod : cawed, rot : wrought may have the same vowel, or may undergo a range of variation from [ä] through [o] to [ȯ]. In r ‑dropping dialects, as that of eastern New England, such leveling may extend not merely to cod and cawed but to cord as well. Card and cord probably are not leveled to any great extent, although, as pointed out earlier, the vowel of card has some degree of rounding in some dialects. For a transcription of the speech of r ‑keepers who level cod : cawed, but not card : cawed, two symbols (as \ä\ and \ȯ\) are necessary in the range between \a\ and \ō\ to take care of the card : cawed distinction. For a transcription of the speech of r ‑droppers who level cod : cawed : cord, two symbols (as \ȧ\ and \ä\ or \ȧ\ and \o\ or \ȧ\ and \ȯ\) are necessary in the range between \a\ and \ō\ to take care of the distinction between card and cod : cawed : cord. For any speakers who may level card : cord as well as cod : cawed, only one symbol (\ȧ\ or \ä\ or \o\ or \ȯ\) is necessary in the range between \a\ and \ō\.
The leveling of words like cod and of words like cawed and cord is not transcribed after the entry words. We have refrained from transcribing it in order to simplify the transcriptions. Users of the pronunciations are to understand that the symbol \ȯ\ and the symbol \ä\, when the latter corresponds to the letter o in the spelling (as in cod ), or corresponds to the letter a in the spelling and is preceded by the sound \w\ or \y\ and is not followed by the sound \r\ plus a consonant (as in watch, yacht, warrant but not in yarn ), are merely, as far as certain dialects are concerned, two symbols for the same sound.
A further complication is the fact that the first vowel of utter in southern British speech is usually more like the usual American [ä] in otter than the usual American first vowel in utter, so that in an ambiguous context an Englishman's utter is apt to be heard by an American as otter.
\ȧ\
as any of several phonetically different vowel sounds that in the speech of r ‑droppers make card distinct from cad on the one hand and from cod on the other, or from cod on the one hand and from cawed on the other.
In the speech typical of eastern New England and southern England, the vowel transcribed \ȧ\ at card is articulatorily between the \a\ of cad and the \ä\ of cod. The New England vowel is closer to the vowel of cad than is the British vowel. In both eastern New England and southern British speech \ȧ\ is usual or frequent in a small group of words that have come to be known as the ask words. In all but a few words of this group, the vowel, orthographically a or au, is followed by one of the voiceless fricatives \f\, \s\, \th\, or by \m\ or \n\ plus a consonant. They are more often pronounced by American speakers with \a\ or (with a few exceptions, at least for \ai\) \aa(ə)\ or \ai\.
In one kind of r ‑dropping speech heard in the southeastern U.S. and in New York City, the vowel transcribed \ȧ\ at card is the same in quality as the vowel transcribed \ä\ at cod but usually of greater duration (see the statement about exceptions at the section on \ä\ ). Some transcribers treat the greater duration as a prolongation of the \ä\ sound, others as a diphthongization of the \ä\ sound to \äə\. In the South, with the exception of a small area of coastal Virginia, \ȧ\ occurs in the ask words only in conscious imitation of the speech of areas in which its use is natural. Its occurrence in New York City is also chiefly imitative, but aunt and rather have \ȧ\ (or \ä\) with some frequency.
Because any word for which only an \a\ pronunciation is shown in this book may with many New York City speakers have a vowel of a quality like that of eastern New England \ȧ\, this latter vowel is perhaps to be regarded as merely a nonsignificant variant of \a\ in their speech.
In another kind of r ‑dropping speech heard in the southeastern U.S. and in New York City, the vowel transcribed \ȧ\ at card is articulatorily between that of cod and that of cawed. Speakers with whom the \ȧ\ is of such quality do not have \ȧ\ in the ask words. Any New York City speakers who may use in the ask words a vowel of the quality that is transcribed \ȧ\ for other dialects are to be regarded as using a nonsignificant variety of \a\. See the preceding paragraph.
Some transcribers use the symbol \ȧ\ for the vowel of most Scottish speakers in words which in most dialects are pronounced with \a\, \aa(ə)\, or \ai\, as cat, man, bag. This Scottish vowel is of much the same quality as eastern New England \ȧ\ but usually markedly shorter. We use for it the symbol \a\. See the section on \ä\.
See also the section on \ä\.
\aa\, \aaə\
chiefly in what is known as the Middle Atlantic region, an area that has New York City and Washington, D.C. on its perimeter. This vowel occurs in one or more of the following groups of words:
(1) words like bare, fair, and Sayre.
(2) words like Cary, hairy, vary, in which typically the sound \r\ represented by a single r in the spelling is followed by a vowel sound and is preceded by a vowel sound whose orthographic representation is a or a vowel sequence containing a. With speakers who have \aa(ə)\ in these words the members of the following groups are usually or often distinct: Cary, carry, Kerry; hairy, Harry; vary, very.
(3) words like sail, sailor, Haley, which in other dialects usually contain the sequence \āl\ or (when not immediately followed by a vowel sound) \āəl\. With speakers who have \aa(ə)l\ in these words, barely and Bailey may be exact rhymes. \aa(ə)\ in this group of words is reported to be less frequent among cultivated speakers.
(4) most words that have in other dialects the vowel \a\ or \ai\ in the last syllable immediately followed by one of the following consonants:
\b\, \d\, \g\, \j\
\f\, \v\, \th\, \th \, \s\, \z\, \sh\
\m\, \n\
When a suffix is added to such words, the vowel usually remains the same, except that the \ə\ final element is usually lost. Before some of the above consonants or some consonant groups of which one of the above is the first member, \aa\ also occurs in syllables that are not final without the help of a suffix addition. The consonants just shown are classified into three groups as an aid to the memory. The first group consists of voiced stops (\j\ is actually a double sound, \d\ + \zh\, the first constituent of which is a voiced stop), the second group consists of voiced and unvoiced fricatives (\zh\ is not listed because there is no fully English word that ends in \aa(ə)zh\), and the third group consists of two of the three nasal consonants. With speakers who have \aa(ə)\ in these words, the vowel of cab is readily perceived as markedly different from the vowel of cap (although those who do not have \aa(ə)\ may have trouble hearing the distinction); can (“container” or “to put in a container”) may be distinct from the emphatic form of can (“to be able”), and halve may be distinct from the emphatic form of have, in that the first member of each pair has \aa(ə)\ whereas the second has \a\; fad and fared when they are identical in stress may be exact rhymes.
In accord with our system of doubling a symbol to indicate duration, \aa\ is longer than \a\. In one variant it is of substantially the same quality as \a\, in a second and possibly commoner variant it is formed with a higher tongue position than \a\ (approaching the tongue position for \e\), and in a third and comparatively infrequent variant it is \ȧ\ (see section on \ȧ\ ). Sometimes \aa\ is followed by a distinct \ə\ off-glide, indicated by three transcriptions \aa\, \aa(ə)\, \aaə\. When the \ə\ is present, it is the overall length of the diphthong that distinguishes or helps to distinguish the vowel from \a\, but to indicate that what we transcribe \aa\ is longer when the \aa\ is a monophthong than when it is the first part of a diphthong would unnecessarily complicate the transcription. 4
\ää\
as the sound \ä\ when it is appreciably long in duration and is distinct from the shorter \ä\ for the reason that both may occur in the same environment. Thus, the German words Stadt and Staat, in the dialect of German usually regarded as standard, consist of the sounds \sh\, \t\, a vowel of \ä\ quality, and \t\, but the \ä\ in Stadt is not as long as the \ä\ in Staat. In our system the German pronunciations are transcribed \ˈshtät\ and \ˈshtäät\ respectively. When such foreign words are pronounced in an English context by one whose native language is English, the durational difference is often lost.
Although the [ä] ¦ [ää] difference in duration serves to distinguish otherwise identical words, as cod and card, in the r ‑dropping dialects of the southeastern U.S. and New York City, we do not employ [ää] in our symbolization of this difference. In r ‑dropping English speech as a whole there are several values of the vowel spelled ar in card that make that word distinct from cod, and all of these values, including [ää], are subsumed under the symbol \ȧ\.
\a(a)(ə)\
\e(ə)\ is to be understood as a variant when it does not appear in a vocabulary transcription, as at cudbear.
\aan \
as the sound of the ein in the French pronunciation of French enceinte — that is, a lengthened \an \. See section on \n \.
\ään \
as the sound of the ande in the French pronunciation of French bande — that is, a lengthened \an \. See section on \n \.
\a(a)rV\
\erV\ and \ārV\ are to be understood as variants when they do not appear in a vocabulary transcription, as at myaria.
accentuation
see the section on stress marks .
\ai\
articulated as a tense monophthong with the tongue higher than for \a\ or as a diphthong having approximately the beginning and ending limits indicated by the \ai\ transcription. Although using the two-symbol transcription is more convenient than devising another diacritic, the sound is in its less emphatic utterances at least a monophthong.
It occurs before some but not all of the consonants listed in classification (4) in the section on \aa\ and \aaə\; it does not occur in words that belong to the other three classifications there; and it occurs before one consonant not listed in any of the four classifications there — the nasal \ŋ\. The occurrence of \ai\ in the dialect of a native of southern Illinois compares as follows with the occurrence of \aa(ə)\ : \ai\ is almost invariable before \ŋ\ even in syllables far removed from the last but does not occur before \b\ and \j\ and occurs before \d\ only in a few words (bad, glad, mad ), chiefly when they are emphatic. \ai\ occurs before \f\, \v\, \th\, \th \, and \s\ chiefly in final and near-final syllables (as in half, bath, baths, the variant \ˈshaivz\ of shafts, and pass ) and before \sh\ even in syllables distant from the last (as in passionate ); it does not occur before \z\ (in jazz for instance) nor in emphatic have or has to. \ai\ does not occur before \m\ or \n\ when one of these is the only consonant that follows (as in ram, ran, except that man may have \ai\ when it is emphatic, as in the interjectional phrase “Man oh man!”) but it does occur before certain consonant sequences of which \m\ or \n\ is the first member (as in sample, answer ). Before \ŋ\, \ai\ may occur in any syllable whereas \aa(ə)\ appears not to occur before \ŋ\ at all. In some parts of the southern U.S., particularly in uncultivated speech, the incidence of \ai\ seems to be wider than in the speech just described.
Precisely what the phonetic relationship is between \aa(ə)\ and \ai\ is uncertain. If both are diphthongs in their maximum-stress occurrences, then the difference in the ending positions of the diphthongs (\ə\ in one case, \i\ in the other) constitutes a distinction. If both vowels are in some of their occurrences monophthongs, then the difference is not so palpable. Both are fairly long vowels, and length therefore is an unlikely basis of difference. With some speakers \aa\ is of the same quality and tongue height as \a\ but longer, whereas with speakers who have \ai\, that vowel is always higher than the speaker's \a\ during all or a part of its articulation — all if \ai\ is a monophthong, a part if \ai\ is a diphthong. When one adds to this the fact that the \a\ (as in cap ) of many Middle Atlantic speakers is an appreciably different vowel from the \a\ (as in cap ) of speakers in most of the rest of the country, it is obvious that \aa\ must often be a quite different vowel from monophthongal \ai\. The \a\ in cap of many Middle Atlantic speakers is obviously a vowel with considerable \ȧ\ coloration in it. Speakers in other parts of the country use in cap a vowel that is much the same in quality (if not in length) as the monophthongal variety of \aa\ vowel in cast. Before certain consonants and with some speakers, the Middle Atlantic vowel that we transcribe \aa\ may be the same as the vowel that we transcribe \ai\. Since the words in which \aa(ə)\ is used instead of \a\ and \ai\ is used instead of \a\ are far from being always the same, it would cause difficulty to try to make only one of these two symbols do the work of both in a given transcription.
There are even fewer pairs of words that are distinguished solely by the difference between \a\ and \ai\ than there are pairs that are distinguished solely by the difference between \a\ and \aa(ə)\. In the speech of some halve is usually distinct from emphatic have and salvy from savvy in that the first member of each pair has \ai\ and the second \a\. The same is probably true of the pair bad : bade in the speech of many.
\äi\, \ȧi\
in the dictionary we have shown no pronunciation for the present participle of the few verbs which have as a final sound \ä\ or \ȧ\ (as sol-fa ), with the implication that the transcription if shown would be \-äiŋ\, \-ȧiŋ\, \-äēŋ\, \-ȧēŋ\. Since, as stated in the section on \ī\, the first component of the diphthong represented \ī\ is usually \ä\ or \ȧ\ and the second component is or may be \i\, the question arises as to whether the vowel sequence immediately preceding the \ŋ\ of such present participles may not be identical with the sequence in fine. For the reasons stated in the section on \ȯi\, random-listening evidence on this matter is difficult to come by. When the last syllable of a verb ending in \ä\ or \ȧ\ is under primary stress (as in hurrahing ), the \ä\ or \ȧ\ usually seems longer than the \ä\ or \ȧ\ of fine; when the \ä\ or \ȧ\ syllable has less than primary stress (as in some utterances of sol-faing ) the two sequences are more apt to be leveled.
\āl\
words transcribed with this succession of symbols, as sail, alien, often have \aa(ə)l\ instead in Middle Atlantic speech. The \aa(ə)l\ variant is not shown in transcriptions in the vocabulary.
\an \
see section on \n \ .
\än \
see section on \n \ .
anglicization
see section on foreign words .
\ār\
occurs in the deep South and in Scotland in words like barbarian and various, and in Scotland in words like fair. In the dictionary \ār\ is not shown for any except Scottish words of the latter type, and is shown only for common words of the former type.
\ au̇\
as the sound of ow in now, ou in loud, and ou in out, and ou in out, and used by many instead of \al\ in some words when certain labial or velar consonants follow, as in scalp, Ralph, album, valve, talcum. When \au̇\ is not an alternative to \al\, the first part of the diphthong may have a range from unrounded \ä\ to \e\, the \e\ occurring especially in dialect. The in-between varieties with \a\ or \ȧ\ are commoner, and it is difficult to be sure whether one is hearing \a\ or \ȧ\ unless there is prolongation of this part of the diphthong, as in the speech of some areas. \ȧ\ or an equivalent is the symbol usually employed for the first part of the sound, but fully as good a case can be made for \a\. There may be prejudice against the use of \a\ because of the exaggerated length of this sound in the diphthong often heard in styles of Southern speech regarded by many as substandard. Such length is not a necessary concomitant of the sound in this diphthong.
In a coastal area of the southern U.S. extending from Virginia to So. Carolina and in some areas of eastern Canada, still another variety of this diphthong occurs — one in which the first element is \ə\ or \ə̄\ or \ō\. Those who have this variety in their speech have the other also, the first occurring before voiceless consonants (as in out ) and the second before voiced consonants and finally (as in loud, now ). Although the two varieties strike even the most untrained ear as widely different, it is not necessary to have two different symbols since their occurrence is determined by the phonetic context.
When \au̇\ is alternative to \al\, the range of \a\ is no wider than that of \a\ when it is followed by a sound other than \u̇\. Although using \au̇\ in the transcription of one class of words in which there is and of another class in which there is not a wide range of variation is not an ideal procedure, neither are the other two possibilities: (1) using \au̇\ as the alternative to \al\ but \ȧu̇\ or some other variant for words of the type out and loud; this would not as readily convey the information that the diphthong in the two classes may be identical; (2) giving the full range of at least five variants for words like out and loud but only \au̇\ when the variant is \al\. This would have made transcribing too space- and time-consuming.
Although we do not show the possibility in vocabulary pronunciations, with some speakers who have \a\ as the first element of their diphthong in loud, the sequence of sounds that we transcribe \al\ in some words and \au̇l\ in others, depending on the spelling, is probably actually \au̇l\ in both classes. Such pairs as Al : owl, Cal : cowl, Hal : howel, Halley : Howley, Alice Ide : owl aside are sometimes identical (the second member of the \l\-final pairs is subject to the intrusion of a parasitic \ə\ sound after the \u̇\, the first is not). The variety of \l\ used by most Americans outside the South in words such as those cited is known as the dark variety, in which the back of the tongue is in a position for \u̇\ or a neighboring vowel. Since the back of the tongue must reach this position before the tip makes the contact for the articulation of an \l\, the intrusion of an \u̇\ or of a closely related vowel sound seems highly probable.
The use of \au̇\ instead of \al\ before \y\, as in medallion, Italian, belongs chiefly to substandard speech. See section on \l\.
\b\
as in baby, knob.
\ḇ\
as in the Spanish pronunciation of hablar. Whereas for \b\ the lips are in contact and form a closure along their entire extent, for \ḇ\ the lips are close together without closure.
brackets, square
[ ] lightface square brackets are used in this guide instead of slant enclosures to enclose transcriptions that are phonetic rather than phonemic.
British dialect
see section on foreign words .
\ch\
based on English orthography, for the sound \t\ followed by the sound \sh\ in the same syllable, as in chin, pitcher, fixture, Christian. The sounds that come between the two vowels in the words cha-cha (dance) and hotshot can in both cases be regarded as \t\ followed by \sh\ but as being in the same syllable in the first (\ˈchäˌchä\) and distributed between the two syllables in the second (\ˈhätˌshät\). If our transcription for what precedes the first vowel in cha-cha had been \tsh\, the transcription for cha-cha and hotshot would have been \tshä·ˌtshä\ and \ˈhät·ˌshät\ respectively.
For words of the type mention, essential, and provincial, transcriptions of the type \-nshən\ and \-nshəl\ have enjoyed a long tradition in dictionaries. At least in American speech and in common words of this type \t\ almost always intrudes between the \n\ and \sh\, and we transcribe \-nchən\ and \-nchəl\.
comparatives of adjectives
see sections on INFLECTED FORMS, DIVISION OF and on INFLECTED FORMS, PRONUNCIATION OF .
correctness in pronunciation
The term correct pronunciation is often used. Yet it is probable that many who use the term would find it difficult to give a precise and clear definition of the sense in which they use it. As every kind of correctness implies a standard of measurement, so in pronunciation it is intimately bound up with the question of standard pronunciation. It has been stated that there are certain extensive regional types of cultivated English speech that have spread far beyond the area of their local origin, as the southern British or American speech is which r is not dropped. It might be reasonably maintained that it would be incorrect for an educated native of London or Oxford to say \ˈask\ or to pronounce the r in farm, just as it would be incorrect for the midwestern American to say \ˈȧsk\ or \ˈfȧm\; for in both cases it is contrary to the standard which prevails in each region. From the nature of the case, when the essential facts are considered, correctness of pronunciation must be a flexible term. It is perhaps as accurate a definition as can be made to say that a pronunciation is correct when it is in actual use by a sufficient number of cultivated speakers. This is obviously elastic, depending both on knowledge — never accurately ascertainable — of the number of users, and on judgment as to the cultivation of the speakers. Mere majorities, without consideration of historical linguistic background and regional distribution, are not decisive. For example, the fact that more speakers in the English-speaking world habitually use R American than any other single type cannot vitiate the standing of the southern British pronunciation for the educated Englishman.
It has been frequently maintained, and more often assumed, that some single type should be looked upon as solely standard. But this is not the prevailing view of those who are familiar with the essential facts (the detailed differences of pronunciation in the different types, their historical development and relations, the various conditions — chiefly uncontrollable — which in the past have brought certain types into prominence) and who, therefore, possess that breadth of view and freedom from local prejudice that result from acquaintance with those phonetic features, often strikingly different, that have attained to approved usage in the standard types of English and other languages. The following statements are worthy of consideration:
Daniel Jones, M.A., emeritus head of the department of Phonetics, University College, London:
“I have no intention of becoming either a reformer of pronunciation or a judge who decides what pronunciations are ‘good' and what are ‘bad’. My aim is to observe and record accurately, and I do not believe in the feasibility of imposing one particular form of pronunciation on the English-speaking world. I take the view that people should be allowed to speak as they like. And if the public wants a standardized pronunciation, I have no doubt that some appropriate standard will evolve itself. If there are any who think otherwise, it must be left to them to undertake the invidious task of deciding what is to be approved and what is to be condemned” — from English Pronouncing Dictionary, N.Y., 1956, p. xvi.
“I do not consider it possible at the present time to regard any special type as ‘Standard’ or as intrinsically ‘better’ than other types.” … “The term ‘Received Pronunciation’ … is often used to designate this type of pronunciation. This term is adopted here for want of a better. I wish it, however, to be understood that other types of pronunciation exist which may be considered equally ‘good’ ” — from An Outline of English Phonetics, 8th ed., N.Y., 1956, p. 12.
A. Lloyd James, M.A., late University reader in Phonetics, School of Oriental Studies, London; linguistic adviser to the British Broadcasting Corporation; and secretary of the B.B.C. Advisory Committee on Spoken English:
“The listener who writes to ask the ‘correct way’ of pronouncing a word quite evidently assumes that there is a ‘correct way’. In all these queries and criticisms there is implied the idea of a standard pronunciation. We have a standard yard, a standard pound weight, a standard sovereign, and a standard pint. The yard does not vary from Aberdeen to Plymouth, and the pint pot contains as much in Mayfair as in Bethnal Green. Unfortunately speech is not capable of rigid measurement, and there is no standard of pronunciation. Pronunciation varies from district to district, from class to class, from character to character, in proportion to the local, social, or moral difference that separates them. … It is quite evident that we are not entitled to conclude that there is one standard pronunciation, one and only one right way of speaking English. There are varieties that are acceptable throughout the country, and others that are not” — from Broadcast English, I., 2d ed., London, 1931, pp. 11 f.
Samuel Moore, Ph.D., late professor of English in the University of Michigan, and editor of the Middle English Dictionary:
“When we consider all the varieties of English spoken by those who are admitted to speak ‘good English’ in the different British colonies and in different parts of the United States, we must recognize that there is still no Standard Spoken English in any strict sense of the term. In every part of the English-speaking world some type of spoken English, that which is used by the educated and superior class within the community, is considered ‘good English’, as contrasted with the ‘Vulgar English’ and local dialects spoken by other classes of the community. If we use the term Standard Spoken English at all we must recognize that it is merely a convenient way of speaking of the various kinds of ‘good English’ that are current in various parts of the English-speaking world” — from Historical Outlines of English Phonology and Morphology, 1929, p. 114.
George Philip Krapp, Ph.D., late professor of English in Columbia University:
“A sufficient definition of the term standard will perhaps be found in the statement that speech is standard when it passes current in actual use among persons who must be accounted as among the conservers and representatives of the approved social traditions of a community” — from The English Language in America, N.Y., 1925, II, 7.
Edward S. Sheldon, late professor of Romance Languages in Harvard University, and editor of etymologies for Webster's International, 1909, and New International Dictionaries, Second Edition, 1934:
“The so-called standard language is not a fixed and infallible standard, but is itself constantly changing with the course of time, and is different in the different places where it is spoken” — from “What is a Dialect?” Dialect Notes, I, 287.
Otto Jespersen, late professor of English in the University of Copenhagen:
“Our chief concern will be with the normal speech of the educated class, what may be called Standard English” — from Essentials of English Grammar, N.Y. [1933], p. 16.
The question, what degree of uniformity exists in the various cultivated types of English speech, depends on what differences one chooses to emphasize, and what to ignore. At present all cultivated types, when well spoken, are easily intelligible to any speaker of English, and there is a very large percentage of practical identity in the speech sounds used. For example, it has been found that, in a thousand words from The Legend of Sleepy Hollow as they would be pronounced by a native of Rochester, N.Y., and a native of London, respectively, there are 125 words that would differ in the use of \r\, 36 in the sound of o as in not, 11 in the sound of a as in ask, and 4 in the use of the secondary accent. These four differences may be looked upon as the differences of pronunciation in the two types most noticeable to the average person, and one of these, the short o, might often pass unnoticed. In most other respects the same phonetic symbols would be used to represent the two types.
Since the establishment of orthoepy as a feature of English dictionaries, the standard assumed has been considerably changed, and has of necessity been made broader. While usage is still and must always be the standard, it is no longer the usage of a particular locality, since the pronunciation of no one locality can now claim admitted precedence. Nor can the pronunciation of any one person, or group of persons, be taken as a standard for all, for such pronunciation is in some cases more advanced, in others more archaic, than the average. Orthoepists of former generations are authorities for the present generation only in so far as their work agrees with good present usage.
The standard of English pronunciation, then, so far as a standard may be said to exist, is the usage that now prevails among the educated and cultured people to whom the language is vernacular; but, as shown above, since somewhat different pronunciations are used by the cultivated in different regions too large to be ignored, we must admit the fact that uniformity of pronunciation is not to be found throughout the English-speaking world, though there is a very large percentage of practical uniformity.
The function of a pronouncing dictionary is to record as far as possible the pronunciations prevailing in the best present usage rather than to attempt to dictate what that usage should be. In so far as a dictionary may be known and acknowledged as a faithful recorder and interpreter of such usage, so far and no farther may it be appealed to as an authority.
In the case of diverse usages of extensive prevalence, the dictionary must recognize each of them.
There is a constantly increasing body of technical terms which, being more often written than spoken, are often called “book words”. For many of these no accepted usage can properly be said to exist, and their pronunciations must be determined on the analogies of words more often spoken, or according to the accepted rules of pronunciation for the languages from which they are derived.
\d\
as in dried, deduce. See section on \d·\ immediately following.
\d·\
when one starts from pause and utters the word die, the consonant is produced by breaking a closure produced by the tongue against the front of the palate, to the accompaniment of vibration of the vocal cords. When one starts from pause and utters the word tie, the consonant is produced by breaking the same closure but without vibration of the vocal cords. The basic difference between the two sounds is that one is voiced and the other is voiceless, and when hearing conditions are not unfavorable, the two can usually be easily enough distinguished without the help of context. In the pronunciation of the dd of ladder and of the tt of latter used by some speakers (British, most consistently), the same readily heard distinction is made. Most U.S. speakers, however, do not use closure (stop) consonants in the words ladder and latter but a voiced flap consonant and the two words cannot be distinguished out of context. This flap consonant is either very similar to or identical with the sound often heard from British speakers for the r of very, the spelling veddy often being used by Americans to mimic this pronunciation.
The voiced flap consonant appears to be normal for d(d) in American speech when a vowel immediately precedes and a vowel or \ə l\ immediately follows, regardless of the stress of the two syllables (ladder, tidal, parody, made-up, grade A, adorn ). It also appears to be normal for t(t) in American speech in the same environment when the second vowel is unstressed (latter, title, parity ) and when the second word is stressed and t(t) ends a word or a word within a word (at Omaha, pick it up, great A, butt end, separatism ). In words in which t(t) is not word-final, a vowel with primary stress precedes, and a vowel with secondary stress follows, both the flap and the voiceless stop (\t \) are quite common (veto, Hittite, Mithridates ). Exceptionally, the flap is fairly frequent for the first t of potato (and is, of course, usual for the second t ).
Some speakers seem to level d(d) and t(t) before \ə n\, in pairs like ridden : written. Such leveling is not as familiar to us and has not been shown in our transcriptions. The leveling sound before \ə n\ is probably a voiced stop rather than a flap.
Most transcribers use a special symbol for the flapped sound. We have employed another approach, however. As previously pointed out, when one starts from pause and pronounces die or tie, the beginning sound is a stop, voiced for die, voiceless for tie. The voiced flap sound does not occur finally before a pause but a typical occurrence of it is for word-final, and hence syllable-final, t(t) before a vowel. The flap that occurs for d(d) and t(t) in American English is always a single sound that follows a vowel and precedes a vowel or syllabic consonant (the \r\ that occurs in r ‑keeping speech in a word like forty is more vowel than consonant — see the section on \r\ ). Vowels and syllabic consonants are syllable nuclei, and there is accordingly a syllable boundary between what precedes and what follows the voiced flap sound. It has seemed altogether practicable to treat the alveolar voiced stop \d\ as a syllable-initial \d\ (transcription \·d\) and to treat the alveolar voiced flap as a syllable-final \d\ (transcription \d·\). Since both the flap and the voiced alveolar stops are heard for intervocalic d(d), the syllable divisor · is not used in the transcription of such words. Since a voiced stop is not used for t(t) before a vowel or \ ə \l\, the flap pronunciation of such words is always indicated by placing the · after the \d\. For all t(t) words that are commonly pronounced with the voiced flap in American English, a pronunciation with the voiceless alveolar stop is usual in some dialects (as British) and occurs with more or less frequency, especially for emphasis, as an alternative pronunciation. Examples of our transcription of d(d) and t(t) before a vowel or \ ə l\:
ladder \ˈladə(r)\ (=\ˈlad·ə(r)\ or \ˈla·də(r)\)
latter \ˈlad·ə(r), ‑atə(r)\ (\ˈla·ə(r)\ is not a normal pronunciation)
editor \ˈedəd·ə(r), ‑ətə(r)\
competitor \kəmˈped·əd·ə(r), ‑etətə(r)\ (this formula is to be understood as also covering a permutation like \-ed·ətə(r)\)
adapt \əˈdapt\ (=\ə·ˈdapt\ or \əd·ˈapt\)
attack \əˈtak\
For uncommon t(t) words, usually only the \d·\ variant is shown in the vocabulary. \t\ is always to be understood as an acceptable variant of \d·\.
Some observers who grant the leveling of the consonant in pairs like ladder : latter insist that the pairs are nevertheless distinct in that the vowel preceding the consonant is longer for the d(d) than for the t(t) word. We have not been sufficiently convinced of such a distinction to show it.
When t is initial in a word with an unstressed vowel or first vowel and follows without pause a word ending with a voiced consonant, or when in the same word t precedes an unstressed vowel and is preceded by a voiced consonant which in turn is preceded by an unstressed vowel, the t is probably often a stop \d\ (not \d·\), as in a dog to watch, hang together, seventy.
dissimilation
see the vocabulary definition. \r\ dissimilation by R speakers (see section on R ) is the most frequent kind in English. When the \r\ affected is preconsonantal, or word-final and therefore potentially prepausal, it is necessary to use the labels R and −R in order to make it clear that the omission occurs in R speech. A mere transcription \sə(r)ˈprīz\ at surprise, for example, would justifiably be interpreted as meaning that \sərˈprīz\ is the R and \səˈprīz\ the −R pronunciation, whereas actually \səˈprīz\ is frequent R as well.
Dissimilation of \l\ is probably next most frequent after \r\ in English, probably accounting, for example, for the frequent pronouncing of only the second l of ophthalmologist and Guadalcanal and of only the first l of Wilhelmina. (The last two words are not included in the vocabulary.)
ditto mark
see 2.8.1, 2.8.2 in “Explanatory Notes ”.
double hyphen
(=) see 2.7.1, 2.7.2 in “Explanatory Notes ”.
doubling
the doubling or repeating of a symbol indicates a sound of greater duration than is indicated by the symbol standing single. In the variants \ˈmanlē\ and \ˈmaanlē\ for manly, \aa\ is a vowel of greater duration than \a\; in \ˈshtäät\, German pronunciation of German Staat, and \ˈshtät\, German pronunciation of German Stadt, \ää\ is a vowel of greater duration than \ä\; in \ˈhīnə̇s\ highness and \ˈfīnnə̇s\ fineness, \nn\ is a consonant of greater duration than \n\. In the case of vowels, there may be a slight difference of quality in addition to the difference in duration but it is not necessary to transcribe this too: it can be regarded as a by-product of the lengthening.
Because of the practices of English orthography and of other pronunciation alphabets, a word of warning is necessary about certain of these doubled symbols that will be found in transcriptions of foreign words and phrases. \ee\ is a vowel having a quality like that of the e in red but of greater duration, not a vowel having the quality of the ee in reed; \ōō\ and \ȯȯ\ are long vowels having a quality like that of the o in coerce and the o in sort respectively, not vowels having the quality of the oo in boot or foot.
\e\
as in bet, bed, and the first syllable of merry; in some American speech, the vowel of the first syllable of marry and the vowel or the core vowel of words like bear and bare.
Some speakers who have \a\ or \aa\ in most words of the type bear, bare, have \e\ in a small number of words of this type (among them care, chair, scare, scarce, emphatic where) in which the preceding sound is articulated with the tongue high in the mouth. Apparently by assimilative influence, the \a(a)\ is replaced by the higher \e\.
Some speakers have \a\ or \aa\ in emphatic there but \e\ in emphatic their and in theirs.
In words of the type barbarian, various, the same speakers have \e\ in some and \a(a)\ in others. Some appear to interchange \e\ and \a(a)\ in the same word. In the deep South and in Scottish, \ā\ occurs in these words. For uncommon words, we have usually transcribed \a(a)\. \e\ and \ā\ are always to be understood as variants of this \a(a)\.
\ē\
as in beat, bead, and the first syllable of beady, and the unstressed vowel used by most American speakers, and apparently also by most Canadian and Australian speakers, in the second syllable of lady, cities, and as a prevocalic in piano, serious, meander. In some parts of the southern U.S., \i\ appears to prevail in these unstressed syllables, and British phoneticians transcribe \i\ although reporting that the British vowel is sometimes \e\ (vowel of bet) or something approaching it. \i\ is favored by some who, while admitting that \ē\ better represents the quality of the usual American vowel, have misgivings that \ē\ in transcriptions may foster a tendency to overpronounce (i.e., to pronounce an \ē\ with full secondary stress).
To hold down the number of variants in transcriptions, only \ē\ has been shown for the i of piano and serious and the first e of meander. It may be, as some transcribers appear to believe, that \ē\ is used in such words by speakers who have \i\ in lady, but \i\ is surely frequent enough to justify the blanket recognition here of \i\ as an always-to-be-understood variant of unstressed \ē\.
\ee\
used only in the transcription of words from foreign languages; of long duration and with a quality like that of the e in English red, not like that of ee in English reed.
\ēr\
occurs in the deep South and in Scotland in words like serial, serious, weary, and in Scotland in words like hear, mere, veer.
It is not shown for any except Scottish words of the latter type, and is shown only for common words of the former type.
\ eu̇\
as a frequent variant of \el\ when certain consonant sounds follow, as in elk, elm, help, twelve. The use of \eu̇\ instead of \el\ before \y\, as in hellion, rebellious, belongs chiefly to substandard speech. See section on \l\.
\f\
as in five, traffic, puff.
foreign words
for foreign words borrowed unchanged into English and italicized by some writers to signify that they are regarded as falling short of complete naturalization, we nevertheless often show in the vocabulary a pronunciation somewhat anglicized. Even an approximation to some foreign sounds is difficult or impossible for the average speaker of English who does not have a speaking knowledge of the foreign language, and even when a sound close to the foreign sound exists in English it may be passed over in favor of one that is usual in the same environment in English words. Thus the Italian name Garibaldi is usually pronounced by English speakers with the first vowel as in garrison and the third as in bald, although the vowel of garb is closer to the one vowel that occurs in both syllables in Italian. The speaker who is sufficiently conversant with the phonetics of a foreign language to be capable and desirous of using the full foreign pronunciation is unlikely to use an English dictionary as a source, and we have accordingly often not given the foreign pronunciation even as a variant. In general we give foreign pronunciations when known usage has shown them or similar pronunciations to be within the capabilities of educated English speakers inexpert in the foreign language, or when no anglicization is known to us to be established and the foreign pattern is such that there is insufficient analogy to make an anglicized pattern safely predictable.
For the same reasons we also show a standard-English pronunciation pattern for words or parts of words belonging to British dialect, including Scottish. Thus the vowel of clat is shown as \a\, the vowel of fat in standard English, and the vowel of darg is shown as \ä\, one of the vowels of bargain in standard English, although the two words have the same vowel in much British dialect. It seems pointless to give only \a\ for words like fat that belong to standard English as well as to British dialect and to give some other vowel for a word like clat that belongs exclusively to British dialect and that rhymes with fat in that dialect. A dictionary key that tells the speaker of standard English who is completely ignorant of British dialect that clat has the vowel of fat provides that consultant with a pronunciation that in his ignorance of dialect he would be well advised to use; a speaker of the British dialect Scottish who finds \a\ for clat in a dictionary and finds fat cited as a key word for \a\ is not misled; a speaker of standard English who knows something about Scottish knows that his vowel in fat is not Scottish dialect's vowel in fat, and also knows what vowel does occur there instead of his vowel.
French words
fully French pronuciations are shown without any stress marks, in accord with the usual practice of transcribers of French.
\g\
as in get, got, tiger, big.
\ḡ\
as for g in Spanish luego; a non-English voiced fricative sound, made with the tongue in approximately the position for \g\ in English go but without closure.
glottal stop
a phonetic effect (symbol \ʼ\) produced by a complete closure in the throat (more specifically, a closure of the vocal cords) that makes it impossible for breath to issue from the lungs. It is the same phenomenon that bottles up the breath before a cough or before the clearing of the throat. In speech it may occur between silence and a sound, between two sounds, or between a sound and silence. In some languages (as standard English, with the exception of a few interjections such as \ʼmʼm\ or \ʼəʼə\ or no), the glottal stop occurs only as an accidental, in some cases almost inevitable, sound, and adds no significance to the context in which it occurs: if it were omitted the meaning of the context would not be altered. Thus, this sound often occurs between \ə\ and a following vowel sound (as in ultra-atomic, ultraism) or before the vowel in a vowel-initial word that is under emphasis (as “Eggs!” or, with an intonation of annoyance or disbelief, “Eggs?”). In speech usually regarded as nonstandard and not recorded in this book, the glottal stop in certain contexts replaces \d·\ or \t\ of standard speech, as in \ˈwȯʼə(r)\ for water. Omission of the stop in such cases might cause misunderstanding if the context were ambiguous enough, since \ˈwȯə(r)\ is a pronunciation of war. In the type of German speech usually regarded as standard, a glottal stop occurs with great frequency before or at the beginning of words stressed on the first syllable and having no consonant letter preceding the first-syllable vowel; it is usually retained when the word follows another member in a compound. Such a glottal stop is symbolized by some transcribers but not by others. Whether it is phonemic or not, it is symbolized in this book in fully German transcriptions of words in which it occurs because its absence hinders ready comprehension by a German listener. We believe that a certain number of departures from strict phonemicity are advisable in the sporadic fully foreign pronunciations.
‑gu-, ‑qu-; division of words containing
for the end-of-the-line division of words like leaguing and cataloguer and piquing, in which gu is pronounced \g\, qu is pronounced \k\, and a following e is dropped before a syllable-increasing English suffix is added, see section on inflected forms, division of .
When the pronounced vowel that follows the u is not an English suffix or a part of one, the most satisfactory division between the two syllables is immediately preceding the g or q, as in belea·guer, li·quor, che·quered.
\gy\
in at least three coastal areas of the South — Virginia, South Carolina, eastern Texas — a \y\ sound often intrudes between \g\ and certain or all of the members of the vowel series from \ē\ to \ä\. Thus, garden is often \ˈgyȧdə n\ in Charleston. These variants are not shown in this book.
\h\
as in hat, ahead; in the anglicized pronunciation of Spanish-derived words, \h\ is the usual pronunciation before a stressed vowel of a sound that in some dialects at least of Spanish is \ḵ\ — for example, for the x of Don Quixote.
\hw\
used for the wh of whet by speakers whose whet and wet are not pronounced the same. Some phoneticians regard \hw\ not as two sounds but as one, a voiceless \w\. With most American speakers whet and wet differ in pronunciation but with most southern British speakers they do not.
\i\
as in bit, bid, here, hear.
\ī\
as in try, light, guide, aisle. \ī\ is a diphthong, not a single sound, with heaviest stress on the first element. Its beginning position ranges from the position for the \ȧ\ heard in ask in eastern New England to the position for the unrounded \ä\ heard from most U.S. speakers for the vowel of hot, heart; its ending position ranges from that for monophthongal \ā\ to that for \i\. In eastern Virginia and in an area of Canada having Toronto as its metropolis, the position of the first element before voiceless consonants is approximately that of the \ˈə\ of nut. In the southern U.S., the second part of the \ȧi\ or \äi\ variety of the diphthong may disappear, and finally and before voiced consonants, less often before voiceless consonants, the pronunciation may be simply \ȧ\ or unrounded \ä\, as in \ˈwȧvz\ or \ˈwävz\ for wives.
See section on \äi\ .
incomplete pronunciations
many of the pronunciations shown in this dictionary are for a part of the entry only and the user is to seek the missing part somewhere else in the dictionary. Two of the most frequently used indicators that a sound symbol or a sequence of sound symbols is not a complete pronunciation for the whole of a boldface entry are the ordinary single hyphen and the vertical bar (for the latter see the paragraph on | and 2.6 of “Explanatory Notes ”). A hyphen at the left-hand side of a roman-type item within reversed virgules means that something is missing at the beginning (\-līn\); a hyphen at the right-hand side means that something is missing at the end (\ˈkōd-\); a hyphen to both left and right means that something is missing at both beginning and end (\-tȯr-\). If the missing part is the pronunciation of a word or of words pronounced at its or their own alphabetical place in the vocabulary, the missing part should be sought there.
law of ti·ti·us \-ˈtētsēəs\ [missing part at law and of ]
gresh·am's law \ˈgreshəmz-\ [missing part at law ]
Otherwise the missing part is to be supplied from the corresponding part of a variant pronunciation that appears before a preceding punctuation mark within the same reversed virgules, or from a pronunciation for another spelling that precedes in the same or another entry.
¹sa·line \ˈsāˌlēn, ‑līn\ [substitute \ˈsā\ for the hyphen]
cod·i·fy \ˈkädəˌfī, ˈkōd-\ [substitute \əˌfī\ for the hyphen]
sen·a·to·ri·al \¦senə¦tōrēəl, ‑tȯr-\ [substitute \¦senə¦\ for the first hyphen and \ēəl\ for the second]
anal·y·sis \əˈnaləsə̇s\ n, pl analyses \-əˌsēz\ [substitute \əˈnal\ for the hyphen]
¹or·gan·ic \(ˈ)ȯ(r)¦ganik, ‑nēk …
²organic \“\ …
or·gan·i·cal \-nə̇kəl, ‑nēk-\ …
or·gan·i·cal·ly \-k(ə)lē, ‑li\ [for the first hyphen at organical substitute \(ˈ)ȯ(r)¦ga\ from ¹organic; for the first hyphen at organically substitute \(ˈ)ȯ(r)¦ganə̇, ‑nē\ from ¹organic and organical ]
We usually repeat something from the preceding pronunciation to indicate where in this pronunciation that which is different in the following pronunciation is to serve as a replacement. Often both something to the left and something to the right of the variant part is repeated. Thus in \ˈsāˌlēn, ‑līn\ the repeated \l\ and \n\ show that \ī\ may replace the \ē\ between those two characters. A locator repeated may be something other than a sound symbol, as a stress mark or a parenthesis.
per·sist \pə(r)ˈsist also ‑ˈzi-\
cre·do \ˈkrē(ˌ)dō, ˈkrā(-\
When we use a consonantal locator on one side of a vowel variant and no locator on the other side, it is to be understood that the vowel adjacent to the locator in a following pronunciation replaces the vowel adjacent to the locator in a preceding pronunciation and is not inserted after that vowel. Thus in \ˈkōdəˌsēz, ‑äd-\ for codices at codex, \-äd-\ is to be read \ˈkäd-\ not \ˈkōäd-\.
Variants of the vowel of a monosyllable or of a sound whose symbolization differs from that of the variant in that one has a diacritic and the other does not or has a different diacritic are sometimes shown without a locator on either side.
bask \ˈbask, ‑aa(ə)-,-ai-,-ȧ-\
broomy \ˈbrümē, ‑u̇-\
hu·chen \ˈhükən, ‑ḵ-\
Printing the last three variants for bask without spaces between is merely a space-saving device that we often employ when two or more variants of each other are printed in succession.
If no hyphen or vertical bar appears at the beginning of a pronunciation, the pronunciation for the first part of the word is complete and nothing is to be supplied ahead of what is shown; if no hyphen or vertical bar appears at the end of a pronunciation, the pronunciation for the last part of the word is complete and nothing is to be supplied at the end of what is shown.
¹good-bye or good-by \gu̇dˈbī, gədˈbī, gəˈbī, ˈbī [the last variant, \ˈbī\, is a pronunciation for the entire word and not for only the part bye or by. ]
²pretty \…, before ‘near’ often ˈpru̇t or ˈprit\ [\ē\ or \i\ is not to be supplied at the end of the last two variants. The pronunciations intended are \pru̇tni(ə)r\, etc.]
A user who is familiar with the pronunciation practices of the preceding edition of this work should observe that the giving of variants differs in the two in that in this book no attempt is made to always have a variant begin at a syllable beginning and end at a syllable ending.
inflected forms, division of
when no division is shown for an inflected form, a form with zero inflection or a form inflected by the addition of an ending that does not add a syllable may be divided at any point where a division is shown in the inflectional base:
²mul·let → pl mul·let or mul·lets
in·ter·vene → in·ter·vened → in·t·er·venes
If the ending adds a syllable and the spelling of the base does not change, a division may be made between the two components, as well as at any point at which a division is shown in the base:
church → church·es
con·strain → con·strained → con·strain·ing → con·strains
ap·proach → ap·proached → ap·proach·ing → ap·proach·es
re·tort → re·tort·ed → re·ing → re·torts
²stout → stout·er → stout·est
In a syllable-increased form in which the final consonant of the inflectional base is doubled, a division is made between the doubled consonants:
re·but → re·but·ted → re·but·ting → re·buts
When both of two identical consonants immediately preceding a syllable-increasing ending belong to the inflectional base, a division is made after the second consonant:
bluff·ing
For variants like conn/con a syllable-increased form can be divided in either of two ways:
conn also con, conn·ing or con·ning
In a syllable-increased form in which a final e of the base is dropped before the ending, a division is made between the letter that preceded the e and the ending:
rate → rat·ed → rat·ing
glue → glu·ing
plague → plagu·ing
pique → piqu·ing
gro·tesque → gro·tesqu·er → gro·tesqu·est
In syllable-increased forms like those in the last three lines, in which gu or qu, with u silent, appears immediately before the ending, some prefer to divide immediately before the g or q if it is not immediately preceded by a short vowel or, in the case of g, by n (haran·guing would suggest the substandard pronunciation \həˈraŋgiŋ\):
pi·quing
pla·guing
grotes·quer
For plurals identical in form with the singular, or formed by adding ‑s or ‑es, usually neither pronunciation nor division is shown. For other plurals pronunciation and division are shown, except that (1) divisions common to both plural and singular are usually omitted and (2) the pronunciation and division of the last part of a compound whose last part is an independent English word must usually be sought at the entry for the independent word. The last or the only centered period in a plural is the last point at which a division should be made in it:
ep·i·the·li·um → pl epithe·lia
The plural may be divided like the singular except that no division should be made after that between the e and l.
The centered period between two consecutive vowels in the plural supersedes a differently placed period between the same two vowels in the singular. In ge·nus, pl gen·era a division is not to be understood between the e and n of the plural as well as between the n and e. It is true that divisions such as ge·nera do occur, especially in British publications, but our policy is to show only one division between two consecutive vowels. A plural of the type not needing division and pronunciation does have pronunciation when it is the first variant plural (as at ²virtuoso where virtuosos is first and virtuo·si second). When a plural of this type is the second variant plural, pronunciation and division are omitted (as at tibia where tibi·ae is first and tibias second).
inflected forms, pronunciation of
The pronunciation of any inflected forms not pronounced is the pronunciation shown at the entry for the inflectional base plus the pronunciation or one of the pronunciations shown for the inflectional ending. These ending entries are, for nouns,¹‑s, ¹‑es ; for verbs, ¹‑ed, ¹‑ing, ³‑s, ²‑es, ²‑est, ¹‑eth ; for adjectives, ¹‑er, ¹‑est . The definitions at some of these contain information that will facilitate the ascertainment of the pronunciation for the entire word.
‑ing
in vocabulary words formed by attaching this ending to a verb, usually no pronunciation is shown, whether the inflected form is a main entry or is a run-on entry, if the pronunciation of that part of the word preceding the ‑ing is the same as the pronunciation of the base verb. When such is not the case, usually the only pronunciation shown for the ‑ing is \-iŋ\ or, for common words entered with hanging indention, \-iŋ, ‑ēŋ. It is to be understood, however, that in some dialects or under some circumstances the ‑ing may be otherwise pronounced. See the entry ¹‑ing in the vocabulary for a detailed list of variants.
iŋ
when stressed, as in sing, forefinger, \āŋ\ and perhaps less often \eŋ\ are variants especially in the southern U.S. These variants are not shown. When unstressed, as in running, \ēŋ\ is a frequent variant and is shown for words that are common.
\ir\
, \iər\, \iə\ in words of the type of inferior, hear, pierce, variants occur that are not shown in the vocabulary or that are not shown for uncommon words.
Words of the type of inferior are shown in the vocabulary as containing the sequence \ir\ followed by a vowel. In the deep South they may contain \ēr\ instead of \ir\, and the \ēr\ variant is to be understood for uncommon words for which it is not shown. In southern British speech all words of this type usually have \ə\ between the \i\ and \r\, and some may be pronounced with \yē\ instead of \iə\.
Words of the type of hear are shown in the vocabulary as having \ir\ or \iər\ in R speech and before a vowel in −R speech, and as having \iə\ in −R speech elsewhere. Words of the type of pierce are shown as having \iə\, with no \r\ following, in −R speech. In some words of both types \yē\ may occur instead of \iə\ in southern British speech. The same variant occurs in a very few words in which the part of the orthography that answers to \iə\, \yə\ in the pronunciation contains no r, as in the thea fragment of theater. 5
\ iu̇\
as in one pronunciation of the il in film, milk. See section on \l\.
\j\
as in jug, badge, agile. Some linguists treat the sound of the italicized letters in the words cited as the sound \d\ followed by the sound \zh\ in the same syllable, others transcribe it with a single symbol.
\k\
as in kick, pach yderm, bacchic.
\ḵ\
as for ch in German siech and Bach; a non-English voiceless fricative sound, made with the tongue in a range of positions from approximately that of the \k\ in English keep to that of the \k\ in English cool but without closure.
With particular reference to German, it was formerly the usual transcriptional practice to have two symbols for this range of sounds, one symbol for the front varieties and another for the back varieties. It is now common practice, however, to regard a single symbol as adequate to the phonemic transcription of the entire range of sounds, and the symbol \ḵ\ is so used in the transcription of the comparatively small number of words for which a full German pronunciation, or something approaching it, is given. Whether or not a single symbol is adequate for transcribing German for Germans, a single symbol suffices in this book, in which any word or phrase borrowed unchanged in spelling from a foreign language is treated as an item occurring in English context. Whatever variety of \ḵ\ may come naturally in a particular word to an educated speaker of English addressing another speaker of English is unlikely to seem out of place in English context. In German the use of one symbol instead of two presents the problem that certain pairs of words (very small in number, apparently) which when spoken in isolation are distinguishable are transcriptionally identical (Pfauchen, pfauchen, both \ˈpfäu̇ḵən\ but with a front \ḵ\ in the first and a back \ḵ\ in the second). If, as it appears may be the case for all except the intervocalic occurrences of \ḵ\, a simple set of rules can be drawn up that will fix its front or back position on the basis of phonetic environment alone, an expediential device for transcribing its quality intervocalically is to treat the front variety as syllable-initial, the back variety as syllable-final, and to employ such transcriptions as \ˈpfäu̇·ḵən\ Pfauchen, \ˈpfäu̇ḵ·ən\ pfauchen, \äḵ·ˈālis\ Achelis.
\ky\
in at least three coastal areas of the South — Virginia, South Carolina, eastern Texas — a \y\ sound often intrudes between \k\ and certain or all of the members of the vowel series from \ē\ to \ä\. Thus, card is often \ˈkyȧd\ in Charleston. These variants are not shown.
\l\
as in leaf, loot, police, allude, feel, fool. Articulations of \l\ vary widely, from the clear l in some pronunciations of leaf to the dark l of fool. In both sounds the tip of the tongue is in contact with the teethridge. For the clear l the top of the tongue is convex; for the dark l the top of the tongue is concave and at the back is in the articulatory position for \ü\ or a neighboring vowel. The clearest l in American speech is not as clear as in some foreign languages, and some phoneticians treat all occurrences of \l\ in the speech of most Americans as dark. The almost invariable practice of all transcribers, whatever their views, is to use only one symbol for all varieties in English. In the U.S. the clearest \l\'s occur in the speech of some Southern speakers, whose pronunciation of the intervocalic \l\ in words like salad, Alice, willing contrasts sharply with that of Northern speakers. The difference between the two can be shown by the transcriptions \ˈsa·lid\, \ˈa·lis\, \ˈwi·liŋ\ for the Southern speakers and \ˈsal·əd\, \ˈal·əs\, \ˈwil·ēŋ\ for the Northern. It is possible that \ˈsau̇l·əd\ and \ˈau̇l·əs\ (in which \a\ is the vowel of apt ) would be proper transcriptions for the dark l articulations of the first two words and that Alice and owl us would be rhymes with some speakers. Such transcriptions have not been shown in this book.
When the vowel articulation preceding an \l\ is one that pulls the back of the tongue forward of the dark l articulatory position, a very clear \ə\ may result while the tongue is in transition between the two positions. Words like ale, eel, and oil, in which the last vowel represented by the spelling is front, often have as much of an \ə\ as betrayal, perigeal, or loyal, especially when a pause or a consonant follows. Before a pause or a consonant, pairs like reel : ideal, trail : betrayal, vile : vial, oil : royal may be exact rhymes. The pronunciation with epenthetic \ə\ is shown as a variant for such words. The back vowel \u̇\ in a word like owl is so far fronted when the first half of the diphthong has the quality of the \a\ in pat that a clear \ə\ frequently intrudes, and the variant with \ə\ is shown in this book. Between other back vowels and \l\ the \ə\ has not been shown although the \ə\ sometimes appears after \ü\ when that vowel is fronted by the articulation of a preceding tongue-front consonant, as in schedule, mule.
\u̇\ often occurs, in some dialects at least, instead of \l\ before the lip consonants \p\, \b\, \m\, \f\, \v\, and before the back-of-the-tongue consonant \k\, especially when these consonants are word-final — that is, only the back half of the articulation of a dark l occurs, the contact between tongue tip and teethridge being absent: help, bulb, elm, self, twelve, elk. With some Southern speakers, help, self, and twelve have nothing whatever corresponding to an l.
Less often than in the class of words just mentioned, \u̇\ occurs instead of \l\ when \y\ immediately follows, as in million, hellion. The \u̇\ variant is not shown before \y\ in this book.
\lə \
as in one pronunciation of French peuple. Like \ə l\, \lə \ is a single sound differing from \ə l\ in being voiceless rather than voiced and in not adding an extra syllable, the \pœplə \ pronunciation of peuple being a single syllable. The \ə \ after the \l\ is always parenthesized to indicate the usual pronunciation when a vowel-initial word follows without pause. Thus peuple anglais is \pœ plän glā\, three syllables, with \pl\ as in French plaisir. Compare section on \ə \.
\lch\
as in belch, gulch; sometimes \lsh\ in U.S. speech and more often so in British speech.
‑lds usually no pronunciation is shown for noun and verb inflected forms produced by adding ‑s to a base in ‑ld, as fields, welds. The pronunciation \l(d)z\ is to be understood.\lü\
as in lute, lewd, clue, ablution, absolute, revolution; that is, words in which the orthographic representation is not loo and which are mostly from Latin and Greek. In words like the first three, the variant \lyü\ (=\lyü\ or \liü\) appears to be frequent in southern British speech but is rare in U.S. speech and is not shown. In words like the last two, in which the \l\ is intervocalic and the sequence \lyü\ can easily be syllabicated \l·yü\ instead of \·lyü\, the variant with \y\ is less rare in the U.S. and is shown after an also.
\ly \
as gli in Italian figlio and ll in the Castilian Spanish pronunciation of olla; the sound can be approximated by trying to pronounce \l\ while the tip of the tongue is held behind the lower front teeth.
\m\
as in maim, hammer, nymph; the usual articulation of \m\ is with the lips in contact to form a closure but when an \f\ or \v\ sound immediately precedes or follows the closure is made by the lower lip against the upper front teeth, as in nymph, triumph, triumvir, Hoffman(n) and the pronunciation \ˈsevə m\ for seven.
\mə \
as in one pronunciation of French prisme; \mə \ is a single sound differing from \ə m\ in being voiceless rather than voiced and in not adding an extra syllable, the \prēsmə \ or \-ēzmə \ pronunciation of prisme being a single syllable. The \ə \ after the \m\ is always parenthesized to indicate the usual pronunciation when a vowel-initial word follows without pause. Thus prisme oblique is \prēz mȯ blēk\ or \prēz mȯ blēk\, three syllables, with \m\ as in French smille or mille. Compare section on \ə \.
\n\
as in known, manly, enrage, tenth.
\n \
indicates that the preceding vowel is pronounced through the nose, that is, with the velum lowered and the nostrils open at the back. Vowels are not infrequently so uttered in American English but the nasality is an accident and a nasal vowel does not function as something different from the same vowel without the nasality. In some languages, however, as French and Portuguese, the difference between a vowel without nasality and the same vowel with nasality may make two different words of two sequences of sounds otherwise identical. French has four nasal vowels and they are often exemplified by the four-word phase un bon vin blanc (“a good white wine”), which in our symbols would be transcribed \œn bōn van blän \.
In a transcription like \kōōn t\ at comte, in which \ōō\ indicates a sound the same in quality as \ō\ but of greater duration (not a vowel with the quality of oo in English boot ), the diacritic \n \ applies to both preceding characters.
‑nds
usually no pronunciation is shown for noun and verb inflected forms produced by adding ‑s to a base in ‑nd, as friends, sends. The pronunciation \n(d)z\ is to be understood.
ng
see section on \ŋ\ .
\ny \
as gn in French agneau and Italian bagno and ñ in Spanish cañón. The sound can be approximated by trying to pronounce \n\ while the tip of the tongue is held behind the lower front teeth.
\ŋ\
as in hang \ˈhaŋ\, hanger \ˈhaŋə(r)\, anger \ˈaŋgə(r)\, singer \ˈsiŋə(r)\, finger \ˈfiŋgə(r)\, linger \ˈliŋgə(r)\. The sound of the ng in hang and of the n in anger is a single sound not the sound \n\ followed by the sound \g\, although the transcription \ng\ for this single sound is used in some of the smaller members of this series to avoid the use of characters that are not letters of the ordinary English alphabet.
\ō\
as in bone, snow, coerce; \ō\ has a strong tendency to become diphthongal, with \u̇\ as a second element, when it is in a position of emphasis, as when it is word-final, under primary stress, and before a pause. In southern British speech and in some American speech, especially in the Philadelphia area, the diphthongization is more frequent and the first element is often \ə\ or \ə̄\. The symbol \ō\ is to be understood to cover all these variants.
In a group of some fifty words, chiefly monosyllables, and their derivatives, transcribed with \ō\, some New England speakers, chiefly rural, have a vowel that has been described, and that apparently is articulated, in more than one way but that sounds to most ears unaccustomed to it like the vowel of cut. At least some New England speakers make homophones of cut and coat. The users of this vowel constitute only a small fraction of the English-speaking world and appear to be on the decrease, and so this variant is not shown.
In a small area that includes the cities of Charleston and Savannah, words transcribed with \ō\ followed by a consonant often have \ōə\ instead, as in \ˈkōət\ for coat. This variant is not shown.
\ȯ\
as in corn, saw, all, cause. In the southern U.S., there is often a diphthongal variant (not used in this book) for at least all of the key words cited except corn. In this variant the first element is much the same as the monophthong that occurs with other speakers in these words and the second element is \u̇\ or even \ü\.
\œ\
as in French bœuf, German Hölle; \œ\ can be approximated by pronouncing \e\ with moderately rounded lips.
\œ̄\
as in French feu German Höhle; \œ̄\ can be approximated by pronouncing \ā\ with strongly rounded lips.
\œn \
see section on \n \ .
\ȯi\
as in coin, boy. In the southern U.S. and chiefly before a consonant in the same word, the second element is sometimes lost or replaced by \ə\. Random listening for misapprehension of two-morpheme items as one-morpheme and vice versa is not as productive of evidence for these pairs as it is for certain other pairs whose identity is disputed — e.g., for latter : ladder and a name : an aim. Even for these common items, highly ambiguous contexts are infrequent, and there are no or few words in ‑oing in English and but few in ‑awy. With most Americans speaking at normal tempo occurrences of \ȯ\ followed by \i\ without pause are phonetically identical, regardless of whether the two components belong to the same morpheme or to two. Whether, however, such sequences are to be regarded as one syllable or two is a more difficult matter. But strawy and the last part of destroy are identical, and coin and boy are more disyllabic than monosyllabic. In the speech of New York City the first half of the diphthong \ȯi\ is overlong.
In transcriptions of the full German pronunciation of German words, \ȯi\ is used for the vowel sequence of words like neu,
whose second member in German is usually a lip-rounded vowel more properly transcribed \ȯ
\ or \ȯœ\.
\ōn \
see section on \n \ .
\ōō\
used only in the transcription of words from foreign languages; of long duration and with a quality like that of the o in English coerce, not like that of oo in English boot.
\ȯȯ\
used only in the transcription of words from foreign languages; of long duration and with a quality like that of the o in English sort, not like that of oo in English foot or boot.
\ōōn \
as \ōō\ nasalized, as in French com te. See sections on \ōō\ and on \n \ .
\ōr\
as in board, boarder, glory; many speakers have this sequence in their speech only when the \ō\ and the \r\ belong to separate morphemes, as in pro rata and the \ə\-less pronunciation of grower. With these speakers, in other items \ȯ\ occurs instead of \ō\. For words common in the literary language, both the \ō\ and \ȯ\ variants are shown. For other words, usually only the \ō\ variant is shown and the \ȯ\ variant is to be understood. Thus, only \-ōr-\ is shown at auctorial and \-ȯr-\ is to be understood, as can be ascertained by examining the transcription at the entry for a common word (like editorial ) for which \-ōrēəl\ is the first variant shown; only \-fō(ə)r\ is shown at gonophore and \-fȯ(ə)r\ is to be understood, as well as \-fōə\ and \-fȯ(ə)\ for r ‑droppers, as can be seen by examining the transcription for a common word like semaphore.
In words like glory and glorious, in which a vowel follows the \ōr\, the pronunciation of many speakers in the deep South, whose vowel is \ō\, is appreciably different from the pronunciation of others in the South and from that of speakers in other parts of the country who have \ōr\ rather than \ȯr\, in that with the first group the \r\ is more consonantal than with the second group, whose \r\ is in the nature of the second half of a diphthong. The first variant may be transcribed \ˈglō·ri\, the second and probably more common variant \ˈglōr·ē, ‑ōr·i\.
\ȯr\
as in moral, horrible, in which a vowel follows the \r\. Both \ȯr\ and \är\ occur in these words in U. S. speech. \är\ strongly predominates in the South, and is frequent along the Atlantic coast in the North. \är\ strongly predominates in British speech, where the \ä\ is usually somewhat lip-rounded and different from the \ä\ of most U. S. speech. \ȯr\ probably prevails in the U. S. and is shown first for nearly all words for which both variants are shown. For words of this group not common in literary English, usually only the \ȯr\ variant is shown but the \är\ variant is to be understood. For a few such words in which the \r\ is or may be followed by \ō\, as borrow and sorrow, many have \är\ who have \ȯr\ in other words of the group. The pronunciation for words like borrow is shown as \är also ȯr\. Dissimilation has been suggested as an explanation of the difference. \är\ is not to be understood for \ȯr\ in words in which the vowel that follows the \r\ is that of a suffix freely addable to English words. \är\ is quite normal for abhorrence and abhorrent, but if it occurs for abhorring or abhorrer it probably does so from the analogy of a word like abhorrence. The analogy of horror, for example, for which \är\ is a normal variant, may sometimes produce \är\ in abhorrer.
\p\
as in pay, lip, upper.
phonemicity
our endeavor has been to make the transcriptions of fully English items phonemic. Because of the inevitability, however, that they will not in every respect agree with every evaluator's ideas of phonemicity, and because we have deliberately not striven for phonemicity in our transcription of certain nonanglicized pronunciations, we enclose our transcriptions in virgules having a slant the opposite of the virgules conventionally employed for phonemic transcription. In the comparatively small number of nonanglicized pronunciations of words from foreign languages that occur in this dictionary of the English language, we have transcribed a variant that is purely allophonic in the foreign language (for example, the [ŋ] variant of \n\ in Spanish and Italian) if the variation is phonemic in English.
plurals
see sections on INFLECTED FORMS, DIVISION OF and on INFLECTED FORMS, PRONUNCIATION OF .
present participles
many present participles hve two pronunciations, one that is the pronunciation of the base verb or of one form of it plus the pronunciation shown in the vocabulary for ‑ing, a second with one syllable less.
¹flick·er \ˈflikə(r)\ … flickering \-k(ə)riŋ\ [i.e., the present participle is \ˈflikəriŋ or \ˈflikriŋ\]
fat·ten \ˈfatə n\ … fattening \-t(ə )niŋ\ [i.e., the present participle is \ˈfatə niŋ\ or \ˈfatniŋ\]
When such a variation occurs, at all but wholly dialect or very rare verbs the present participle is subentered and pronounced. Usually only \-iŋ\ is shown at the end but it is to be understood that other variants of \-iŋ\ may occur; see ¹‑ing in the vocabulary. When more than one identically spelled verb is entered the present-participle pronunciation may be shown only for the common one or ones or for the first.
principal parts of verbs
see sections on INFLECTED FORMS, DIVISION OF and on INFLECTED FORMS, PRONUNCIATION OF .
punctuation marks
variants are usually separated by a comma in the vocabulary but sometimes by a semicolon. See the paragraph on variants.
‑qu-: division of words containing
see sections on ‑gu- and on INFLECTED FORMS, DIVISION OF .
R
used in vocabulary pronunciations and in this “Guide” as a label for the speech of those who always have the sound \r\ before a consonant or pause when r or rr in the spelling provides justification, except for occasional loss by dissimilation as in the first syllable of surprise, The term “r ‑keeper” has been used in a few places for a speaker of this type of speech, to supply the need for an antonym of “r ‑dropper”.
−R
used in vocabulary pronunciations and in this “Guide” as a label for the speech of those who have no \r\ sound before a consonant or pause. Our use of this label ignores speakers who in this position sometimes have \r\ and sometimes do not in the same class of words, or who usually have \r\ only in a limited number of classes from which educated usage sanctions its omission. Thus some speakers usually have \r\ before consonant or pause only in syllables of the type of spurn or spur. Speakers whose omission of \r\ is maximal and those whose omission is only partial are both known as “r ‑droppers” but in this “Guide” the term is restricted to the former.
\r\
as in rid, arouse, merry as pronounced by all speakers of standard English, and as in one pronunciation of carbarn, lizard, murder.
Words such as the last three are not uniformly pronounced. Most Americans have for these r 's a sound which, though it may usually differ appreciably from the prevocalic second sound in crow, is distinct from any other sound in the language and can be symbolized by \r\. The \r\ and the vowel sound that answers to a letter preceding the r in the orthography may be articulated simultaneously, as in purr and perhaps also in par, and some transcribers prefer a single symbol for the double articulation. Favoring the use of \r\ preceded by a vowel symbol, however, is the fact that the articulations, usually or often simultaneous, may alternatively be successive, particularly under emphasis, and that the two-symbol transcription parallels the transcription for items like dear, whose corresponding articulations are only or usually successive. Other speakers (as in the southeastern United States, eastern New England, New York City, and southern England) with varying degrees of regularity do not make in words like carbarn, lizard, murder the articulation that we transcribe \r\. The sum total of what they do instead is commonly called r ‑dropping, a term that is misleading to the extent that it may suggest that syllables in which some have \r\ and some do not are identical except that with the latter one member of a linear series has been removed, with consequent durational shortening. Trying to gauge the comparative duration of what R and −R speakers respectively say for what is between the \k\ and \d\ of card runs into complications. A vowel of the quality of that in card, if it occurs and is of substantially the same quality in the speech of both in items in which the spelling does not have r, may not be of the same duration in the two dialects. Further, it may be difficult or impossible to dissociate the vowel and the \r\ in R speech: the \r\ may be simultaneous with the whole or the last part of the vowel, or may be initiated after the vowel begins and continue after the vowel ends. In most classes of words with postvocalic r in the orthography things happen in the pronunciation that make these words as distinct from maximally similar words spelled without r as in R speech. The relationships in −R speech between the vowels used for items containing postvocalic r in the spelling and the vowels used for items not containing such an r are:
(1) The pronunciation of a spelling item containing the letter r may be identical with the pronunciation of an item not containing r:
tort and taut may be identical, with [ȯ] or [ȯȯ] (long [ȯ]) or [ȯə] for both or and au. Card and cod are often identical (as in the speech of New York City), with [ää] for both ar and o, [ää] occurring for the latter because of the tendency of certain final consonants, as voiced stops, to lengthen a preceding vowel. The consonants that when word-final tend to lengthen a preceding vowel and to produce such leveling are \b\, \d\, \g\, \j\, \m\, \n\, \sh\. Although published statements on −R speech take little note of the possible leveling of ar and o as in cart and cot for the opposite reason that the vowel for the r ‑containing item has the shortness normal for the o of an item like cot, in certain contexts at least that are most favorable to vowel shortness (distance in syllables from the end of a word as in parsimony, stress weaker than primary as in arcade ) such leveling may occur. To hold down the amount of detail in our vocabulary transcriptions, we have not indicated there the levelings here discussed.
Pairs like manners : mannas \-nəz\, scapulars : scapulas \-ləz\, rushers : Russia's \-shəz\, goer : Goa \-ōə\, Ballard : ballad \-ləd\ are identical.
The following pairs may be identical, some chiefly in eastern New England and southern British speech, both members containing [ää] or [ä] ([ä] here, in square brackets, transcribes a vowel that is between \ä\ and \a\ in quality; within reversed virgules in vocabulary transcriptions, \ȧ\ stands for both [ää] and [ȧ]; see the section on \ȧ\ ): farther : father, spars : spas, aren't : aunt, arms : alms, carve : calve, farced : fast, parsed : past.
(2) The pronunciations of both items with and items without r in the spelling may contain the same sounds in the same sequence but be distinct because the vowel is longer in the items with r.
Pairs like cart and cot so differ with a high degree of consistency in the South and in New York City, pairs like card and cod with much less consistency. See (1) preceding.
Pairs like bird and bud may so differ, and the difference may be shown by the transcriptions [ˈbəəd] and [ˈbəd] respectively. The words more often differ in that bird has [ə̄] instead of [əə], [ə̄] indicating a vowel differing in quality from that of bud. Within reversed virgules in transcriptions, \ə̄\ stands for both [ə̄] and [əə].
(3) Items with and items without orthographic r may differ in that in the pronunciation of the former the vowel common to both has \ə\ following:
beard \ˈbiəd\: bid \ˈbid\, erred \ˈeəd\: Ed \ˈed\, court \ˈkōət\ coat \ˈkōt\, poorly \ˈpu̇əlē\: pulley \ˈpu̇lē\, sired \ˈsīəd\: side \ˈsīd\, scour \ˈskau̇ə\: scow \ˈskau̇\, coir \ˈkȯiə\: coy \ˈkȯi\. Some speakers may frequently have \ȯə\ in words like tort but usually only \ȯ\ in words like taut. Because of a tendency for English vowels to shorten in proportion to their distance in syllables from the end of a word, it was decided at the beginning of transcription to show \ȯə\ as well as \ȯ\ for r words of this class only in the ultima and penultima — e.g., in norm and normal but not in normalize or normality. The precedent, however, of words that show \ȯə\ only or alternatively in at least all stressed syllables and of other words that show only \ȯ\ in all syllables has made for some inconsistency in the transcriptions.
The \ȯə\ of some transcribers and the \ȯȯ\ of others (see 2) may sometimes be merely a difference of interpretation of the same phonetic entity. We do not show a transcription \ȯə\ for words of the type taut, law. For −R dialects in which tort and taut are both \ˈtȯət\ the variant \ȯ\ should be regarded as having the value \ȯə\. Some transcribers might regard the entity transcribed \ää\ in (1) as in pairs like card : cod, cart : cot, as \äə\ instead.
The Middle Atlantic vowel symbolized \aa\ is often followed by \ə\ in monosyllables or in the last syllable of words without r in the spelling, with leveling of pairs like paired : pad, as indicated by the variants \ˈpaaə\ at pair and \ˈpaa (ə)d\ at pad.
In drawled speech, especially common in the South but not here transcribed, the vowel \ə\ frequently occurs between certain vowels and a following consonant, as in \ˈbiəd\ for bid. Whether pairs like beard : bid are leveled in the South and if so to what extent is a question that needs further investigation.
(4) An item containing r may differ in −R pronunciation from all items not containing r in that a vowel is used for the r item that lacks \r\ quality and that is different from any other the speaker uses. Thus the vowel \ə̄\ is limited to words like bird, German, except as it may occur as a replacement for \œ\, \œ̄\, \œœ\, \œ̄œ̄\ in words originating in a foreign language, as Goebbels, Goethe. The diphthong \əi\ is used for the ir of bird by many speakers in the New York City area and in the deep South.
(5) The speech of some is partly −R and partly R. Speakers whose speech habits were formed near a boundary between −R and R areas sometimes have and sometimes lack \r\ in words of the same class (card, cart, hearth ) or even in the same word. Many speakers have in words of the class bird, dirt, mirth the same pronunciation as R speakers — \ər\ — but a basically −R pronunciation in other classes of words.
What happens before a vowel in −R speech to a word in which \r\ is the last symbol shown in transcriptions for R speech needs some discussion.
When before an initial vowel in a following word there is a pause whose correlate in the written language is a punctuation mark, there usually is no \r\ at the end of a preceding word when there would be no \r\ if a consonant followed: Why should I care? Others don't.
When there is no pause between two such words, −R speech may have an \r\ or may not: a door opened. When there is no \r\ a glottal stop may occur between the two words. Such a glottal stop is not transcribed in this dictionary.
When the vowel that follows is a suffix or the first sound of a suffix, \r\ regularly occurs in all standard −R speech: starry, starring.
A word like dear is not pronounced with an \r\ at the end in −R speech when a consonant or pause follows (“dear friend”, “Dear? Eighty dollars!” ), but is often pronounced with an \r\ when a vowel word follows without pause (dear experience ). A word like idea also is not pronounced with an \r\ when a consonant or pause follows, but like dear frequently does have an \r\ before a closely following vowel with many speakers who pronounce an \r\ in dear in the same situation, in spite of the fact that idea has no r in the orthography. Such an \r\, called “intrusive r ”, is frequent in the northeastern United States and southern England but is rare or nonexistent in the southern United States. It usually follows \- ə\, \ȯ\, \ȧ\, or \ä\ and may occur before a vowel in a following word or, perhaps less often and with greater deprecation from some listeners, before a suffixed vowel: idea of importance, law and order, Omaha and Lincoln, drawing, withdrawal, sol-faing. Our transcriptions have taken notice of intrusive r probably only at idea, where the variant with \r\ was mentioned both because it seemed a distortion to omit it from a complement of variants so sizable and because idea has come to be regarded as something of a shibboleth for the occurrence of intrusive r. Two kinds of parenthesizing of an \r\ at the end of a transcription will be found: “(r)”, with a parenthesis on each side, and “(r”, with no closing parenthesis:
dod·der \ˈdädə(r)\
¹bar \ˈbär, ˈbȧ(r\
In transcriptions with two parentheses the form with r is the R pronunciation and in some environments the −R pronunciation also, the form without r is the −R pronunciation in other environments. In that fraction of a transcription that follows a comma or semicolon and that ends with r preceded but not followed by a parenthesis, the form with as well as the form without r is limited to −R speech (as \ˈbȧr\ as well as \ˈbȧ\); the form preceding the comma or semicolon (as \ˈbär\) is the R pronunciation and with −R speakers whose \ȧ\ is to be construed as [ää] is alternative to [ää] before a vowel.
In words that rhyme with spur and are common enough to call for maximum variant coverage, the R and −R pronunciations are labeled.
In a pronunciation like that at
fear \ˈfi(ə)r, ‑iə\
\ˈfi(ə)r\ is the R pronunciation in all environments and the −R pronunciation in some, \ˈfiə\ is the −R pronunciation in other environments.
To provide a degree of clarification of the pronunciations of such words impossible in most of our transcriptions, to enable the consultant to be sure of our intent as to the pronunciation of inflected forms not transcribed, and to enable the consultant to supply the −R pronunciation of uncommon words for which that pronunciation is not shown, we transcribe in detail below the infinitive, past, and present participle of one verb of each of the classes into which verbs are divisible on the basis of the pronunciation of the last syllable.
fear \R, −R + suffixal vowel, some −R + following-word vowel ˈfir, ˈfiər; −R + consonant or pause, some −R + following-word vowel ˈfiə\ feared \R ˈfi(ə)rd, −R ˈfiəd\ fearing \R & −R ˈfi(ə)riŋ\ (the greater the stress or emphasis the greater the likelihood of \-iər(-)\ rather than \-ir(-)\)
bare \R, −R + suffixal vowel, some −R + following-word vowel ˈbar, ˈber, ˈbaar, ˈbaər, ˈbeər, ˈbaaər; −R + consonant or pause, some −R + following-word vowel ˈbaə, ˈbeə, ˈbaaə\ bared \R ˈba(ə)rd, ˈbe(ə)rd, ˈbaa(ə)rd, −R ˈbaəd, ˈbeəd, ˈbaaəd\ bearing \R & −R ˈba(ə)riŋ, ˈbe(ə)riŋ, ˈbaa(ə)riŋ\
bar \R ˈbär; −R + consonant or pause, some −R + following-word vowel ˈbȧ; some −R + suffixal vowel, some −R + following-word vowel ˈbȧr; some −R + suffixal or following-word vowel ˈbär (with some speakers whose ˈbȧ = ˈbää)\ barred \R ˈbärd, −R ˈbȧd\ barring \R ˈbȧriŋ, −R ˈbȧriŋ or ˈbäriŋ
spur \R + consonant ˈspər, +vowel ˈspər·; −R + consonant or pause ˈspə̄, +suffixal vowel ˈspər· also ˈspə̄r, +following-word vowel ˈspər· or ˈspə̄ also ˈspə̄r\ spurred \R ˈspəard, −R ˈspə̄d\ spurring \R ˈspər·iŋ, −R ˈspər·iŋ also ˈspə̄riŋ\ (the centered period means that only one of the two variants that occur for the urr of a word like hurry occurs in spurring; see the section on \¦ər\ ; the variants \ər·\, \ə̄r\ differ in that although both have substantially the same two articulations the articulations are simultaneous or overlapping for \ər·\ but consecutive for \ə̄r\, the latter being of greater duration than the former)
dodder \R, −R + suffixal vowel, some −R + following-word vowel ˈdädər; −R + consonant or pause, some −R + following-word vowel ˈdädə\ doddered \R ˈdädərd, −R ˈdädəd\ doddering \R & −R ˈdäd(ə)riŋ\ (in most words of this class, with unstressed final syllable, a variation occurs for the present participle that it is more convenient to record at the subentry for that form)
war, warred, warring \as for variants with ȯ at store below \
store \R, −R + suffixal vowel, some −R + following-word vowel ˈstōr, ˈstōər, ˈstȯr, ˈstȯər; −R + consonant or pause, some −R + following-word vowel ˈstōə, ˈstȯ, ˈstȯə\ stored \R ˈstō(ə)rd, ˈstȯ(ə)rd, −R ˈstȯ(ə)d\ storing \R & −R ˈstō(ə)riŋ), ˈstȯ(ə)riŋ\
tour \R, −R + suffixal vowel, some −R + following-word vowel ˈtu̇r, ˈtu̇ər; −R + consonant or pause, some −R + following-word vowel ˈtu̇ə\ toured \R ˈtu̇(ə)rd, −R ˈtu̇əd\ touring \R & −R ˈtu̇(ə)riŋ\
tire \R, −R + suffixal vowel, some −R + following-word vowel ˈtīr, ˈtīr; −R + consonant or pause, some −R + following-word vowel ˈtīə\ tired \R ˈtī(ə)rd, −R ˈtīəd\ tiring \R & −R ˈtī(ə)riŋ\
tower \R, −R + suffixal vowel, some −R + following-word vowel ˈtau̇r, ˈtau̇ər; −R + consonant or pause, some −R + following-word vowel ˈtau̇ə\ towered \R ˈtau̇(ə)rd, −R ˈtau̇əd\ towering \R & −R ˈtau̇(ə)riŋ\
\r(·)\
see \ə·r, ər·\ and \r\ . Words that rhyme with stir and that are not common often have \r(·)\ as the final items in their transcriptions, as knur \ˈnər(·)\. The parenthesized centered period means that before a following vowel such words do not have one of the two pronunciations that are common in standard speech for the urr of hurry. For the full range of pronunciations that are heard for a word like knur see \r\ in this “Guide” or the pronunciation of rhyming words in the vocabulary.
\rə \
as in one pronunciation of French lustre.
\rə
\ is a single voiceless sound that does not add an extra syllable, the \l
strə
\ pronunciation of lustre
being a single syllable. The \ə
\ after the \r
\ is always parenthesized to indicate the usual pronunciation when a vowel-initial word follows without pause. Thus lustre antique
is \l
strän
tēk\, three syllables, with \str\ as in French strate.
See section on \ə
\.
r ‑dropper
see section headed −R.
r ‑keeper
see section headed R.
\rr\
in some languages, as Spanish, the difference between a trilled r (made with two or more taps of the tongue) and a single-tap r is phonemic, distinguishing words otherwise identical. In transcriptions of the Spanish pronunciation of words borrowed from the Spanish \rr\ has been used for the trilled r. For languages which have a trilled r only or in which the difference between two varieties is of no significance, \r\ alone is adequate.
\rü\
as in rude, peruse, verruca; that is, words in which the orthographic representation is not roo and that are mostly from Latin and Greek. The variant \ryü\ or \riü\ occurs in such words to a very limited extent in U.S. speech and so is not shown for any of the words listed above. Occasionally when \ˈrü\ and \- r(y)ə\ are gradational variants (as in garrulity, garrulous ), the \ryü\ variant may be less limited in its extent and then is shown.
\s\
as in so, less, lesser. On the showing of \z\ and \s\ variants for the s of words like abstain, teamster, instigate (in which a voiced consonant precedes and a voiceless sound follows), see the section on \z\ .
sandhi
see the vocabulary definition. Only to a very limited extent is account taken in vocabulary pronunciations of the variations that occur at the beginning of words because of the nature of the last sound of a preceding word, or of the variations that occur at the end of words because of the nature of the first sound of a following word. In common words whose last sound is \t\ when the word is pronounced in isolation, we record that before vowels \d·\ usually occurs instead, as at complete. At and and you we record variations that these words undergo as a result of environment. At 'em we record that a preceding word ordinarily \v\-final may before \ə m\ have \b\ instead of \v\. We do not, however, record at words whose final sound is shown as \s\ or \z\ that instead the sound is usually \sh\ or \zh\ respectively when a word following without pause begins with \sh\, \y\, or \zh\, as in horseshoe, the pace you've set, the size you want, this gendarme. We do not record at words whose first sound is shown as \b\, \d\, or \g\ that when a word ending in \s\ precedes without pause these three stops may not differ in pronunciation from the stops respectively transcribed \p\, \t\, and \k\ in spy, stay, sky, with frequent homophony of such pairs as this buy: the spy, this day: the stay, this guy: the sky (compare our transcriptions at disband, disdain, and disguise ). We do not record at words shown as having a given consonant at the end of an unstressed final syllable and at words shown as having the same consonant as their first sound that when such a consonant-final word precedes such a consonant-initial word without pause often only one consonant is articulated, with homophony of such pairs as Asian nights: Asia nights. We do not record at words shown as beginning with unstressed \əl-\ or \il-\, \əm-\ or \im-\, \ən-\ or \in-\ that when certain consonants precede without pause some of the words of the first group may begin \ə l-\ instead (as in not allowed ), some of the words of the second group may begin \ə m-\ instead (as in the help employed ), and some of the words of the third group may begin \ə n-\, \ə m-\, or \ə ŋ\ instead (as in good encyclopedia, stop entirely, and dog encountered respectively). These sandhi variants that we omit from vocabulary entries are more satisfactorily covered by being cataloged in a discussion of phonetics. Most of them happen automatically and usually without any awareness of the speaker that he is not using the pronunciations that dictionary vocabularies usually limit themselves to.
\sch\
as in mischief, exhaustion. Many transcribe only \sch\ in such words but a sequence whose first member is \sh\ rather than \s\, or at least whose first member seems to be closer to \sh\ than to \s\, is frequent when the vowel that precedes is front and is as high as \e\, as in mischief, question, the tongue position for a vowel of this height being closer to the tongue position for \sh\ than for \s\. After other vowels, whose tongue position is less conducive to \sh\, that consonant is less frequent and is not shown as a variant in the vocabulary, as at bastion, combustion, exhaustion.
Scottish
see section on foreign words .
semicolon
a semicolon rather than a comma sometimes separates variant pronunciations. A semicolon does not mean that what follows it is less acceptable or less frequent than what precedes it. It is in general used as a safeguard against regarding a variant or a label as applicable to an item or a part of an item following the semicolon. Examples:
adversary \R ˈadvə(r)ˌserē, −R ‑vəˌs-; ‑ri\ (if a comma had been used instead of the semicolon, the meaning would be that the variant ‑ri is limited to −R speech)
cayuga \kēˈügə, ˈkyü-; kä(y)ü-ˌkīˈ(y)ü-, attrib (ˈ)=¦==\ [the attrib (ˈ)=¦== does not apply to anything before the semicolon]
aficionada \əˌfisēəˈnädə, əˌfēsēəˈ-,əˌfishəˈ-, aˌf-, ‑ȧdə; ‑ˌfēthēəˈnäthə, ‑nȧ-\ (such variants as \i\, \s\, and \sh\ are unlikely when the last consonant in the word is th)
luxembourg or luxemburg \ˈlu̇ksəmˌbu̇rg, ‑bu̇əg also ˈlük- or ‑ˌbərg or ‑ˌbə̄g or ‑ˌbəig; ˈləksəmˌbərg, ‑bə̄g, ‑ˌbəig also ‑bu̇rg or ‑bu̇əg\ (the label also does not apply to ˈləksəmˌbərg, which is probably as frequent as the ˈlu̇k- variant; and construing the also to apply to the last four final-syllable variants would produce a contradiction with the first part of the pronunciation. Use of the semicolon facilitates the presentation of the relationship between the vowel quality of the first and final syllables)
\sh\
as in shy, dish, sure, mission. \sh\ is a single sound, not two, and has in it no \s\ sound or \h\ sound.
\shch\
see section on \sch\ .
‑sia, ‑sian
as in magnesia, euthanasia, Andalusion. For many words ending in ‑sia or ‑sian as many as ten variants occur: \-zhə(n), ‑zhēə(n), ‑shə(n), ‑shēə(n), chiefly Brit ‑ziə(n), ‑zyə(n), ‑zhyə(n), ‑siə(n), ‑syə(n), ‑shyə(n)\. We do not show this range of variants at any entry in the vocabulary. For common words for which we have records we limit the transcriptions to the variants that appear to be most frequent in American speech, and for uncommon words we often show \-zh(ē)ə\ as the two variants that an averaging of the records of commoner words suggests as most likely.
stress marks
see 2.2 in “Explanatory Notes ”; the mark - is treated in a separate part of the “Guide”. The brief discussion of stress in “Explanatory Notes”, while true as far as it goes, omits some details. Actually, words like beneficial (vocabulary stress marking \¦==¦==\), fictitious (vocabulary stress marking \(ˈ)=¦==\), and campaign (vocabulary stress marking \(ˈ)=¦=\) undergo an even wider variation of stress than “Explanatory Notes” shows. Thus beneficial may also be \ˈ==ˈ==\ (as in Beneficial though it is …) or \ˌ==ˌ==\ (as in beneficial diet), fictitious may also be \ˈ=ˈ==\ or \ˌ=ˌ==\ (as in fictitious story ), and campaign may be \=ˈ=\ (as in a hard campaign ), \ˈ=ˌ=\, \ˈ=ˈ=\, or \ˌ=ˌ=\ (as in campaign promises ).
A high percentage of adjectives traditionally transcribed with a primary stress as the last of two or more stresses (as beneficial, scrophulariaceous ), and many adjectives traditionally transcribed with a primary stress on the second syllable but no stress on the first (as fictitious, magnanimous, dissociative ) are marked in the vocabulary with the indicators of variation of stress. (Two successive primary stresses were mentioned as a possible stress variant for beneficial; three successive primaries seem unlikely but three successive secondaries may easily occur, as in the fourth word of one of the scrophulariaceous herbs. ) A high percentage of words belonging to other parts of speech and traditionally transcribed with two or more stresses the last of which is a primary on the last syllable (as campaign, acquiesce, catamaran ) are transcribed with the markings for stress variation because of the frequency with which primary stress recedes in them or is reduced to secondary when another word follows without pause (acquiesce readily, catamaran sailing ). It is possibly a safe generalization to say that there is no word having two or more stresses the last of which is primary when the word is pronounced in isolation that does not in some contexts undergo recession or reduction of the last stress, whatever the part of speech and whichever the syllable that bears the last stress. Thus in the item Smith's explanation, Johnson's explanation, Wilson's explanation — they're all questionable, probably neither of the two stressed syllables in any of the three words explanation would be uttered with primary stress. But such a stress pattern would occur in fewer contexts for a noun that in isolation lacks primary on the ultima than for a noun that possesses it, and our stress marking for explanation is \ˌ==ˈ==\.
For two-member compounds the second member of which is an independent English word the vocabulary usually shows only the pronunciation of the first member followed by a plus sign. If ¦ or (ˈ) occurs in the transcription of the first member, it is to be understood that the compound as a whole is subject to stress variation. When in compounds whose pronunciation is shown by the plus-sign formula the second member normally has secondary stress only, the plus sign is followed by ˌ-.
mac·ro·fauna \¦makrō+\
mac·ro·prism \ˈmakrō+ˌ-\
Many words for which variable stress is indicated are both adjective and noun. If for such a word that has no stress on the last syllable the adjective entry is first, we usually show the pronunciation of the noun following simply as \“\. If the noun entry is first, we usually show for it a pronunciation in which the only primary stress is the last and show variation of stress for the adjective following by means of stress marks and double hyphens. Thus our pronunciation for ¹organic, adj is \(ˈ)ȯ(r)¦ganik, ‑nēk\ and for ²organic, n is \“\. If ¹organic had been the noun instead our pronunciation would have been \ȯ(r)ˈganik, ‑nēk\ and the pronunciation for ²organic, adj would have been \(ˈ)=¦==\. Showing no stress variation for the noun in such cases is an oversimplification: even a noun without stress on the last syllable does sometimes undergo stress recession but much less often than does the adjective.
The nouns that usually have last-syllable primary stress when pronounced immediately before a pause are shown with the marks for recession of stress. If, in words with last-syllable stress, this stress is often secondary rather than primary before a pause their stress may be indicated by two variants separated by a comma.
¹av·oir·du·pois \ˌavə(r)dəˈpȯiz, ˈ===ˌ=\
\sü\
as in sue, suit, assume; that is, words in which the orthographic representation is not soo and that are mostly from Latin and Greek. The variant \syü\ or \siü\ appears to be frequent in southern British speech but is rare in U.S. speech and so is not shown.
superlatives of adjectives
see sections on INFLECTED FORMS, DIVISION OF and on INFLECTED FORMS, PRONUNCIATION OF .
symbol names
the terms bar, one-dot, and two-dot can be used with the name of a character in this way: \ā\ is “bar a”, \th \ is “bar t-h”, \ȯ\ is “one-dot o”, \ü\ is “two-dot u”. Symbols with no modifier are plain: \a\ is “plain a”, \i\ is “plain i” because the dot is not a modifier.
\t\
as in tights, attend, Atlantic. See section on \d·\.
With chiefly substandard speakers in the New York City area the precise phonetic form of the consonant answering to t or tt between a preceding stressed vowel and a following \ə l\ is the glottal stop, as in title, battle (symbol, not shown in the transcription of such words in the vocabulary, [ʼ]). In Scotland the glottal stop is used in the same environment, and before an unstressed vowel as in water, bitter, by a higher percentage of cultured speakers than in New York City.
In words like winter, plenty, gentlemen, in which what corresponds to the nt of the spelling is preceded by a stressed vowel and followed by an unstressed vowel or \ə l\, the \t\ is either feebly articulated or absent in much American speech, pairs like winter : winner being difficult or impossible to distinguish without the help of context. Certainly the usual American pronunciation is in strong contrast to the usual southern British pronunciation, in which there is a strongly articulated, aspirated, distinctly heard \t\ that probably is to be regarded as the first sound of the syllable to which the unstressed vowel or \ə l\ belongs, any \t\ that may occur in the American pronunciation probably being best regarded as belonging to the syllable that contains the \n\. One way of transcribing the difference between the two pronunciations would be \ˈwin(t)·ə(r), ˈwin·tə(r\. Orthoepic commentary usually decries the absence or weakness of the \t\ in such words but the pronunciation is too widespread in all levels of American speech to be ignored. The \t\-less pronunciation is noticed in the vocabulary only in an occasional word for which multiple variants are shown or from which the absence is regarded as apt to be especially conspicuous, as at gentleman, the emphasis on the plural of this word in the formula “Ladies and Gentlemen” at the beginning of an address making the absence quite noticeable. A two-variant pronunciation such as that shown above for winter is to be understood in the vocabulary for all words of this class, with the label “÷” preceding the \-n(t)·-\ variant to signify the deprecation of orthoepists.
With chiefly substandard speakers in the New York City area a pronunciation somewhat like the British often occurs, except that the tongue position is further forward, the tip frequently being against the lower front teeth.
\th\
as in thin, ether. \th\ is a single sound, not two, and has in it no \t\ sound or \h\ sound. The basic difference between \th\ and \th \ is that the first is pronounced without and the second with vibration of the vocal cords.
\th \
as in then, either. \th \ is a single sound, not two, and has in it no \t\ sound or \h\ sound. The basic difference between \th\ and \th \ is that the first is pronounced without and the second with vibration of the vocal cords.
‑ths
plurals of singulars ending in ‑th pronounced \th\ with vowel preceding and without pronunciation in the vocabulary are to be understood to be \ths\, as myths. Plurals with the variants \th z\ and \ths\ (not always in this order) are pronounced in the vocabulary. The \th \ of \th z\ is often weakly articulated; in fact, it may be that in some pronunciations regarded as containing \th \ what is so regarded is an extension of the length of the preceding vowel rather than a consonantal articulation.
\ū\
this symbol, used in most diacritical alphabets for the sounds following the \f\ of few and often for the sounds between the \f\ and \r\ of fury, is replaced in this dictionary by \yü\ and \yu̇\, which are not only phonetically more realistic but also transcriptionally more economical in permitting the showing of two variants by parenthesization, as in \ˈn(y)ü\ for new.
\u̇\
as in pull, wood, injurious. See section on \u̇r\.
\u̇\ is usually shown as the second part or ending position of the diphthong of loud but in most articulations a point this high is not reached, the ending position being closer to \ō\. Other diphthongs with this ending position occur as variants in certain classes of words in which other speakers pronounce a vowel followed by \l\. See section on \l\.
\
\
as in German füllen, hübsch;
\
\ can be approximated by pronouncing \i\ with moderately rounded lips.
\
\
as in French rue,
German fühlen;
\
\ can be approximated by pronouncing \ē\ with strongly rounded lips.
\ü\
as in rule, moon, few \ˈfyü\, union \ˈyünyən\. Compare the section on \yü\.
\ü\ plus unstressed vowel
in vocabulary pronunciations, \u̇\ if not shown is to be understood as a variant of \ü\ in such position, as in fluid, gluey, skua.
\ u̇r\, \ u̇ər\, \ u̇ə\
in words of the type of injurious, furious, tour, pure, variants occur that are not shown in the vocabulary or that are not shown for uncommon words.
Words of the type of injurious are shown in the vocabulary as containing the sequence \u̇r\ followed by a vowel. In the deep South they may contain \ür\ instead of \u̇r\, and the \ür\ variant is to be understood for uncommon words for which it is not shown. In southern British speech all words of this type usually have \ə\ between the \u̇\ and \r\ and may have \ōər\ as a variant; the most common words may also have \ȯr\ or \ȯər\ as a variant, and for one such word at least (injurious ) \ə̄r\ is also shown.
Words of the type of pure are shown in the vocabulary as having \u̇r\ or \u̇ər\ in R speech and before a vowel in −R speech, and as having \·ə\ in −R speech elsewhere. In southern British speech the \ə\ appears to be usually present between \u̇\ and \r\; in all such words a variant \ōə(r\ is heard, and the most common words may also have \ȯə(r\ or \ȯ(r\.
V
used in the formula “usu ‑ād· +V” as at wait; chosen for the formula as being an abbreviation for “vowel” but “+V” is to be interpreted as meaning “+ vowel or ə l following without pause ”.
\v\
as in vote, level, give. Under the assimilative influence of a following \ə m\, the lip-teeth sound \v\ may have as a variant the two-lip sound \b\, as in give 'em, government. See section on \ə m\.
variants
see 2.3, 2.4 in “Explanatory Notes ”. If the consultant knows or has reason to believe that an uncommon word has less than a full complement of the predictable variants that we usually give for common words, he can supply the omitted variants by examining the variants for better-known words that rhyme or the variants for a rhyming part of such words. He will need to observe certain precautions mentioned in this “Guide”. The \əi\ variant shown at a word of the type of bird, for example, does not hold for a word of the type of preferred. See \əi\.
\w\
as in we, sweep, away. See sections on \əw\, \hw\.
\wy \
as hu in French huile or u in French nuit; \wy \ can be approximated by rounding the lips as for \w\ while the tongue makes the articulation for \y\.
x
for words of the type of exact, exult, in which the sounds that correspond to the x of the spelling are preceded by an unstressed and followed by a stressed vowel, the vocabulary usually shows only the value \gz\ for the x. For such words some speakers have \ks\ instead.
\y\
as in yard, yours, European \¦yu̇rə¦pēən\, cue \ˈkyü\, union \ˈyünyən\. See section on \yü\.
\y \
not a symbol for a sound but a diacritic signifying a modification of the sound of the preceding symbol, the modification consisting of articulating the sound while the tongue is in approximately the position for the sound \y\, with the tip back of the lower front teeth. See the sections on \ly \, \ny \, \wy \.
Frequently at the end of the articulation of \ny \ or \ly \ an independent \y\ sound is heard, without anything else accompanying, [ny y], [ly y], but it is usually considered unnecessary to transcribe this off-glide.
The sound \wy \ could with equal logic have been symbolized by a full-size y followed by a superior w to denote lip rounding, but \wy \ is preferred because \w\ is the usual anglicization, as in \swād\ for suede, from a French word pronounced \swy ed\.
\yə\
as in ammunition. Some transcribers prefer \yu̇\ to \yə\ as the transcription in an unstressed medial syllable of u or eu. In normal-tempo speech it is difficult to hear the difference between unstressed \yə\ and \yu̇\. A distinct \yu̇\ is probably more common in British speech than in American and is heard in the latter in emphatic or deliberate utterance. In initial syllables, as in unite and uranium, \yü\, \yu̇\, and \yə\ are all three heard although the vocabulary does not show all three. When such words are pronounced starting from pause, the \yə\ variant is more apt to occur before \r\ than before other consonants, and the vocabulary transcriptions for the initial vowel are not identical for all words.
\yü\, \ yu̇\
these two transcriptions replace the ū of previous editions of this dictionary, \yu̇\ occurring before \r\ as in European and \yü\ elsewhere as in unity. These two transcriptions not only better display the nature of the sounds but also make it possible to show two pronunciations for a word like new simply by parenthesizing, \ˈn(y)ü\, the (y) signifying that new may either be pronounced like the noo- of noose or have between \n\ and \ü\ a sound that does not occur in noose. See sections on \lü\, \sü\.
When certain consonants precede in the same syllable the first component may be [i] instead of [y]. In the variety with [y] the second component has greater stress than the first but in the variety with [i] the first component has stress equal to or greater than that of the second. It is not necessary to show in the vocabulary the variant [iü] or [iu̇], which is merely the variety of \yü\ or \yü\ that occurs under certain conditions.
See section on \yə\.
\z\
as in zone, freezer, raise. Most transcribers treat as \s\ the sound of the s in words like abstain, teamster, instigate, kidskin, brownstone, in which a voiced consonant precedes and a voiceless sound follows. But it is doubtful that there is any difference between most utterances of the ‑mst- of teamster and the ‑m's t- of the team's terrific, of the ‑dsk- of kidskin and the ‑d's k- of kid's kin, of the ‑nst- of brownstone and the ‑n's t- of Brown's tone, in the second member of which pairs transcribers just as regularly treat the sound of the s as \z\. What happens in both brownstone and Brown's tone when they are identical is probably the same thing that phoneticians agree happens to the last sound of Brown's or brows when those words have only pause following, as at the end of a sentence: the last sound starts out voiced, by carryover from the voiced sound preceding, but becomes voiceless.
This devoicing might be shown in minute transcription intended primarily for scholars by the symbol \z\ with a small circle underneath, such a circle being widely used by phoneticians for devoicing, or even by a transcription \brau̇nzstōn\, but for an item like Brown's which has at the end the same sequence of voice and voicelessness a plain \z\ alone is usually employed. For items like abstain, teamster, instigate we show two variants for the s, \z\ and \s\, the latter probably being frequent or usual in emphatic utterance. We provide only stress indications at the vocabulary entries of compounds like kidskin and brownstone, although only \s\ is shown at skin and stone . If transcriptions with sound symbols were shown for such compounds our transcription at kidskin would be \ˈkidzˌkin, ‑dˌsk-\. Compare the section on sandhi.
When the preceding voiced consonant is a nasal, when the next vowel is without stress, and when the s is definitely \s\ and not \z\, the voiceless stop homorganic with the nasal often occurs parasitically before the \s\: teamster \ˈtēmpstə(r)\, monster \ˈmäntstə(r)\, gangster \ˈgaŋkstə(r)\.
\zh\
as in vision. \zh\ is a single sound, not two, and has in it no \z\ sound or \h\ sound.
\zü\
as in presume, resume; that is, words in which the orthographic representation is not zoo and which are mostly from Latin and Greek. The variant \zyü\ or \ziü\ appears to be frequent in southern British speech but is rare in U.S. speech and so is not shown.
1 \ŋ\ is not involved here because it is the only velar consonant other than stops. Consonant successions like \ln\, \rn\, \rl\ in words like sullen, barren, barrel are not considered, for this reason: these successions can be pronounced monosyllabically, as in one pronunciation of kiln, in barn, and in one pronunciation of Carl (see section on rl ), and the usual practice among phoneticians for these two-consonant successions capable of monosyllabic pronunciation (as successions like \tn\ and \bm\ are not) is to transcribe a vowel between them when they are pronounced as in sullen, barren, and barrel, just as \v\ and \d\ are \-vd\ in the monosyllable lived but \-və̇d\ in the disyllable livid.
2 In some other languages, as French, \l\, \m\, and \r\ are in such position pronounced in such a way as to make possible a monosyllabic transition between them and a preceding consonant of the type illustrated. The qualification should also be added that we are speaking of successions of sounds in which the second is in prepausal position: since the desideratum in these considerations is to see what happens to a succession when an attempt is made to attach it to a single vowel, to introduce a word like jitney, in which the members of the succession are susceptible to distribution between a preceding and a following vowel, is to introduce a different criterion.
3 Parenthesizing of \r\ after a vowel in a stressed syllable will in most cases not yield the r ‑droppers' pronunciation.
4 This account is based to a large extent on that of Allan F. Hubbell in his Pronunciation of English in New York City (1950) but he is in no way responsible for our interpretation of his data.
5 Daniel Jones, Outline of English Phonetics, New York (1956), §442a.
AA | Associate in Arts |
AAS | Associate in Applied Science |
ab | about |
AB | Bachelor of Arts |
abbr | abbreviation |
Abd | Abdias |
AB in Th | Bachelor of Arts in Theology |
abl | ablative |
AC | alternating current |
acad | academy |
acc or accus | accusative |
act | active |
A.D. | anno Domini |
adj | adjective |
adv | adverb |
advt | advertisement |
Aeol | Aeolic |
AF | Anglo-French |
Afr | Africa, African |
Afrik | Afrikaans |
Agg | Aggeus |
agric | agriculture |
A.H. | anno Hegirae |
Ala | Alabama |
Alb | Albanian |
alter | alteration |
a.m. | ante meridiem |
AM | amplitude modulation, Master of Arts |
Am or Amer | America, American |
AmerF | American French |
AmerInd | American Indian |
AmerSp or AmSp | American Spanish |
anat | anatomy |
anthrop | anthropological, anthropology |
aor | aorist |
Apoc | Apocalypse |
appar | apparently |
Apr | April |
Ar | Arabic |
Aram | Aramaic |
archeol | archeology |
archit | architecture |
arith | arithmetic |
Ariz | Arizona |
Ark | Arkansas |
Arm | Armenian |
art | article |
assoc or assn | association |
ASSR | Autonomous Soviet Socialist Republic |
Assyr | Assyrian |
Assyr-Bab | Assyro-Babylonian |
astrol | astrology |
astron | astronomy |
ASV | American Standard Version |
at.no. | atomic number |
attrib | attributive, attributively |
at.wt. | atomic weight |
aug | augmentative |
Aug | August |
Austral | Australia |
Av | Avestan |
AV | Authorized Version born |
b | born |
BA | Bachelor of Arts |
Bab | Babylonian |
bacteriol | bacteriology |
BA (J) | Bachelor of Arts in Journalism |
Bar | Baruch |
BArch | Bachelor of Architecture |
bart | baronet |
BBC | British Broadcasting Corporation |
B.C. | before Christ, British Columbia |
BChE | Bachelor of Chemical Engineering |
BCS | Bachelor of Commercial Science |
BD | Bachelor of Divinity |
Bé | Baumé |
BE | Bachelor of Education, Bachelor of Engineering |
BEd | Bachelor of Education |
bef | before |
Bel | Bel and the Dragon |
Belg | Belgian, Belgium |
Beng | Bengali |
BFA | Bachelor of Fine Arts |
biochem | biochemistry |
biog | biographical, biography |
biol | biologic, biological, biology |
BJ | Bachelor of Journalism |
bk | book |
BldgE | Building Engineer |
Blitt or Blit | Bachelor of Letters, Bachelor of Literature |
bot | botany |
Braz | Brazilian |
Bret | Breton |
brig | brigadier |
Brit | Britain, British |
bros | brothers |
BS | Bachelor of Science |
BSA | Bachelor of Science in Agriculture |
BSc | Bachelor of Science |
BSEd or BSE | Bachelor of Science in Education |
BSFS | Bachelor of Science in Foreign Service |
BS in CE | Bachelor of Science in Chemical Engineering, Bachelor of Science in Civil Engineering |
BS in ChE | Bachelor of Science in Chemical Engineering |
BS in Ed | Bachelor of Science in Education |
BS in LS | Bachelor of Science in Library Science, Bachelor of Science in Library Service |
bt | baronet |
Btu | British thermal unit, British thermal units |
Bulg | Bulgarian |
bull | bulletin |
C | centigrade |
cal | calendar, caliber |
Calif | California |
Canad | Canada, Canadian |
CanF | Canadian French |
Cant | Cantonese |
cap | capital, capitalized |
capt | captain |
Cast | Castilian |
cat | catalog |
Catal | Catalan |
caus | causative |
cc | cubic centimeter, cubic centimeters |
CE | Chemical Engineer, Civil Engineer |
Celt | Celtic |
cen | central |
cent | century |
cgs | centimeter-gram=second |
Chald | Chaldean |
chem | chemical, chemistry |
Chin | Chinese |
Chron | Chronicles |
cm | centimeter, centimeters |
co | company, county |
col | colonel |
Col | Colossians |
coll | college |
colloq | colloquial |
Colo | Colorado |
com | common |
comb | combination, combining |
compar or comp | comparative |
conj | conjunction |
Conn | Connecticut |
cons | consonant |
constr | construction |
Copt | Coptic |
Cor | Corinthians |
Corn | Cornish |
corp | corporation |
cos | cosine |
CPA | Certified Public Accountant |
CRL | Canons Regular of the Lateran |
crystallog | crystallography |
cu | cubic |
cwt | hundredweight |
cycl | cyclopedia |
Czech | Czechoslovak, Czechoslovakia, Czechoslovakian |
d | denarius, denarii, penny, pence |
D | Dutch |
Dan | Daniel, Danish |
dat | dative |
D.C. | District of Columbia |
DC | direct current |
DD | Doctor of Divinity |
Dec | December |
def | definite |
Del | Delaware |
dept | department |
Deut | Deuteronomy |
dial | dialect |
dict | dictionary |
dim | diminished, diminutive |
Dlitt or Dlit | Doctor of Letters, Doctor of Literature |
dom | dominant |
Dor | Doric |
doz | dozen |
DP | domestic prelate |
dr | debit, doctor |
DSc | Doctor of Science |
Du | Dutch |
DV | Douay Version |
E | east, eastern, English |
eccl | ecclesiastic, ecclesiastical |
Eccles | Ecclesiastes |
Ecclus | Ecclesiasticus |
ecol | ecological, ecology |
econ | economics |
ed | edition |
EdM | Master of Education |
educ | education, educational |
EE | Electrical Engineer |
EFris | East Frisian |
e.g. | exempli gratia |
EGmc | East Germanic |
Egypt | Egyptian |
embryol | embryology |
emu | electromagnetic unit, electromagnetic units |
encyc | encyclopedia |
Eng | England, English |
engin | engineering |
Eph | Ephesians |
Esd | Esdras |
Esk | Eskimo |
esp | especially |
esq | esquire |
EST | eastern standard time |
Esth | Esther |
esu | electrostatic unit, electrostatic units |
etc | et cetera |
Eth | Ethiopic |
ethnol | ethnology |
eve | evening |
Exod | Exodus |
explan | explanatory |
Ezech | Ezechiel |
Ezek | Ezekiel |
f | following |
F | Fahrenheit, French |
FBI | Federal Bureau of Investigation |
Feb | February |
fem | feminine |
ff | following |
fig | figurative, figuratively, figure |
Finn | Finnish |
fl | flourished |
Fla | Florida |
Flem | Flemish |
FM | frequency modulation |
fo or f | folio |
f.o.b. | free on board |
fps | foot-pound-second |
fr | from |
Fr | French |
freq | frequent, frequentative, frequently |
Fri | Friday |
Fris | Frisian |
ft | feet, foot |
fund | fundamental |
fut | future |
g | acceleration of gravity |
G | German |
Ga | Georgia |
gal | gallon, gallons |
Gal | Galations |
gaz | gazette |
gen | general, genitive |
Gen | Genesis |
geog | geographic, geographical |
geol | geologic, geological, geology |
geom | geometry |
Ger | German |
GI | government issue |
Gk | Greek |
gm | gram, grams |
Gmc | Germanic |
Goth | Gothic |
govt | government |
Gt Brit | Great Britain |
h | hour, hours |
Hab | Habacuc, Habakkuk |
Hag | Haggai |
handbk | handbook |
HE | his excellency |
Heb | Hebrew, Hebrews |
HG | High German |
hist | historical, history |
Hitt | Hittite |
HMS | Her Majesty's Ship, His Majesty's Ship |
hon | honorable, honorary |
hort | horticulture |
Hos | Hosea |
hp | horsepower |
ht | height |
Hung | Hungarian |
Icel | Icelandic |
i.e. | that is |
IE | Indo-European |
Ill | Illinois |
IMF | International Monetary Fund |
imit | imitative |
imp or imper | imperative |
imperf | imperfect |
in | inch, inches |
inc | incorporated |
incho | inchoative |
Ind | Indiana |
indic | indicative |
infin | infinitive |
infl | influenced |
inst | institute, institution, institutional |
instr | instrumental |
interj | interjection |
internat or internatl | international |
interrog | interrogative |
intrans | intransitive |
inv | inversion |
Ion | Ionic |
IPA | International Phonetic Alphabet |
Ir | Irish |
Ire | Ireland |
IrGael | Irish Gaelic |
irreg | irregular |
Isa | Isaiah, Isaias |
ISV | International Scientific Vocabulary |
It or Ital | Italian |
ital | italic, italicized |
Jan | January |
Jap | Japanese |
Jas | James |
Jav | Javanese |
Jer | Jeremiah, Jeremias |
jg | junior grade |
Jn | John |
JND | just noticeable difference |
Jos | Josue |
Josh | Joshua |
jour | journal |
JPS | Jewish Publication Society |
jr | junior |
Jth | Judith |
Judg | Judges |
K | Kelvin (scale) |
Kans | Kansas |
KC | king's counsel |
KCB | Knight Commander of the Bath |
kg | kilogram, kilograms |
kgps | kilogram per second, kilograms per second |
km | kilometer, kilometers |
Kp | kitchen police |
kt | knight |
Ky | Kentucky |
L | Latin |
La | Louisiana |
LaF | Louisiana French |
Lam | Lamentations |
lang | language, languages |
lat | latitude |
lb | pound, pounds |
l.c. | lowercase |
Lev | Leviticus |
lg | large |
LG | Low German |
LGk | Late Greek |
LH | lower half |
LHD | Doctor of Humanitites |
LHeb | late Hebrew |
lieut | lieutenant |
lit | literal, literally, literary |
Lith | Lithuanian |
LittB or LitB | Bachelor of Letters, Bachelor of Literature |
LittD or LitD | Doctor of Letters, Doctor of Literature |
Lk | Luke |
LL | Late Latin |
LLB | Bachelor of Laws |
LLD | Doctor of Laws |
loc | locative |
long | longitude |
LP | long-playing |
lt. | lieutenant |
ltd | limited |
LXX | Septuagint |
m | meridies (L., noon), meter, meters, minute, minutes |
M | monsieur |
MA | Master of Arts |
Macc | Maccabees |
Maced | Macedonian |
Mach | Machabees |
mag | magazine |
maj | major |
Mal | Malachi, Malachias |
manuf | manufacture, manufacturing |
Mar | March |
masc | masculine |
Mass | Massachusetts |
math | mathematics |
MBA | Master of Business Administration |
MBret | Middle Breton |
Md | Maryland |
MD | Doctor of Medicine, Middle Dutch |
Me | Maine |
ME | Middle English |
mech | mechanical |
med | medical, medicine |
MEd | Master of Education |
met | metropolitan |
meteorol | meterology |
Mex | Mexican, Mexico |
MexSp | Mexican Spanish |
MF | Master of Forestry, Middle French |
mfg | manufacturing |
MFlem | Middle Flemish |
MFS | Master of Foreign Study |
mg | milligram, milligrams |
MGk | Middle Greek |
MHeb | Middle Hebrew |
MHG | Middle High German |
Mic | Micah |
Mich | Micheas, Michigan |
mil | military |
min | minor, minute, minutes |
Minn | Minnesota |
MIr | Middle Irish |
misc | miscellaneous |
Miss | Mississippi |
Mk | Mark |
mks | meter-kilogram-second |
ml | milliliter, milliliters |
ML | Medieval Latin |
MLA | member of the legislative assembly |
MLC | member of the legislative council |
MLG | Middle Low German |
Mlle | mademoiselle |
MLS | Master of Library Science |
mm | millimeter, millimeters |
Mme | madame |
Mo | Missouri |
mod | modern |
modif | modification |
Mon | Monday |
Mont | Montana |
MP | member of parliament, military police |
MPA | Master of Public Administration |
MPer | Middle Persian |
mph | miles per hour |
MPP | member of provincial parliament |
MS | manuscript, Master of Science |
MSc | Master of Science |
MS in LS | Master of Science in Library Science |
MSS | manuscripts |
MSw | Middle Swedish |
mt | mount, mountain |
Mt | Matthew |
MusM | Master of Music |
MW | Middle Welsh |
mythol | mythology |
n | noun |
N | north, northern |
Nah | Nahum |
nat or natl | national |
NATO | North Atlantic Treaty Organization |
naut | nautical |
N.C. | North Carolina |
NCAA | National Collegiate Athletic Association |
NCE | New Catholic Edition |
N. Dak. | North Dakota |
NE | northeast |
Nebr | Nebraska |
Neh | Nehemiah |
neut | neuter |
Nev | Nevada |
NewEng | New England |
NewZeal | New Zealand |
NGk | New Greek |
NGmc | North Germanic |
N.H. | New Hampshire |
NHeb | New Hebrew |
N.J. | New Jersey |
NL | New Latin |
N. Mex. | New Mexico |
no | north, number |
nom | nominative |
nonattrib | nonattributively |
nonstand | nonstandard |
Norw | Norwegian |
Nov | November |
n pl | noun plural |
N.S. | Nova Scotia |
NT | New Testament |
Num | Numbers |
numis | numismatic, numismatical, numismatics |
NW | northwest |
NWT | Northwest Territories |
N.Y. | New York |
NYC | New Yorjk City |
N.Z. | New Zealand |
O | Ohio, old |
Obad | Obadiah |
OBret | Old Breton |
obs | obsolete |
OBulg | Old Bulgarian |
OC | Cistercian Order |
O Cart | Carthusian Order |
OCatal | Old Catalan |
occas | ocassionally |
OCorn | Old Cornish |
Oct | October |
OD | Old Dutch |
ODan | Old Danish |
OE | Old English |
OF | Old French |
OFM | Order of Friars Minor |
OFris | Old Frisian |
OFrk | Old Frankish |
OHG | Old High German |
OIcel | Old Icelandic |
OIr | Old Irish |
OIt | Old Italian |
Okla | Oklahoma |
OL | Old Latin |
old-fash | old-fashioned |
OLF | Old Low Franconian |
ON | Old Norse |
ONF | Old North French |
ONorw | Old Norwegian |
Ont | Ontario |
o.p. | out of print |
OP | Order of Preachers |
OPer | Old Persian |
OPg | Old Portuguese |
OPol | Old Polish |
OProv | Old Provençal |
OPruss | Old Prussian |
Oreg | Oregan |
orig | originally |
ORuss | Old Russian |
OS | Old Saxon |
OSB | Order of St. Benedict |
Osc | Oscan |
OScan | Old Scandinavian |
OSD | Order of St. Dominic |
OSlav | Old Slavic |
OSM | Order of the Servants of Mary |
OSp | Old Spanish |
OSw | Old Swedish |
OT | Old Testament |
OTurk | Old Turkish |
OW | Old Welsh |
oz | ounce, ounces |
p | page |
Pa | Pennsylvania |
PA | protonotary apostolic |
PaG | Pennsylvania German |
paleontol | paleontology |
Par | Paralipomenon |
part | participate |
pass | passive |
pathol | pathology |
PC | privy councilor |
Pek | Pekingese |
Per or Pers | Persian |
perf | perfect |
perh | perhaps |
pers | person |
Pet | Peter |
pfs | private first class |
Pg or Port | Portuguese |
PG | preacher general |
pharm | pharmacy |
PhB | Bachelor of Philosophy |
PhC | Pharmaceutical Chemist |
PhD | Doctor of Philosophy |
Phil | Philippians |
Philem | Philemon |
philos | philosophy |
PhilSp | Philippine Spanish |
photog | photography |
phr | phrase |
physiol | physiology |
pkg | package |
pl | plural |
p.m. | post meridiem |
Pol | Polish |
polit | political |
pop | population |
poss | possessive |
pp | pages |
prep | preposition |
pres | present, president |
Pr Man | Prayer of Manasseh |
prob | probably |
pron | pronoun, pronunciation |
prond | pronounced |
pronunc | pronunciation |
protect | protection |
prov | province |
Prov | Provençal, Proverbs |
prox | proximo |
Ps | Psalms |
Psychol | psychology |
PTA | Parent-Teacher-Association |
pub | public |
publ | publication |
rap | rapid |
RD | rural dean |
recd | received |
redupl | reduplication |
ref | reference |
refl | reflexive |
rel | relative |
relig | religion |
rep | republic |
repr | representatives |
resp | respectively |
ret | retired |
rev | reverend, review |
Rev | Revelation |
rev. ed. | revised edition |
R.I. | Rhode Island |
Rom | Roman, Romance, Romanian, Romans |
ROTC | Reserve Officers' Training Corps |
rpm | revolutions per minute |
RR | railroad |
RSV | Revised Standard Version |
rt hon | right honorable |
Rum | Rumanian |
Russ | Russian |
s | shilling, shillings |
S | south, southern |
S.A. | South Australia |
Sam | Samuel |
Sat | Saturday |
SB | Bachelor of Science |
Sc | Scots |
S.C. | South Carolina |
Scand | Scandinavian |
ScD | Doctor of Science |
ScGael | Scottish Gaelic |
sci | science |
Scot | Scotland, Scottish |
S. Dak. | South Dakota |
SE | southeast |
sec | second, seconds, section |
secy | secretary |
Sem | Semitic |
Sept | September |
Serb | Serbian |
Shak | Shakespeare |
sin | sine |
sing | singular |
SJ | Society of Jesus |
SJD | Doctor of Juridical Science |
Skt | sanskrit |
Slav | Slavic |
SM | Master of Science |
so | south |
SoAfr | South Africa |
soc | social, society |
sociol | sociology |
Sol | Solomon |
Soph | Sophonias |
South | Southern |
Sp or Span | Spanish |
specif | specific, specifically |
sp. gr. | specific gravity |
sq | square |
sr | senior |
SSR | Soviet Socialist Republic |
st | saint, street |
stand | standard |
stat | statue |
STB | Bachelor of Sacred Theology, Bachelor of Theology |
STD | Doctor of Sacred Theology |
ste | saint |
STL | Licentiate of Sacred Theology |
STM | Master of Sacred Theology, Master of Theology |
subj | subjunctive |
substand | substandard |
Sun | Sunday |
superl | superlative |
supp | supplement |
Sus | Susanna |
Sw or Swed | Swedish |
SW | southwest |
syll | syllable |
syn | synonym, synonymy |
Syr | Syriac |
Tag | Tagalog |
tech | technical, technological, technology |
technol | technology |
tel | telegraph, telephone |
Tenn | Tennessee |
Tex | Texas |
ThD | Doctor of Theology |
theol | theological |
Thess | Thessalonians |
Thurs | Thursday |
Tim | Timothy |
Tit | Titus |
Tob | Tobias, Tobit |
Toch | Tocharian |
Toch A | Tocharian A |
Toch B | Tocharian B |
trans or transl | translated, translation |
Tues | Tuesday |
Turk | Turkish |
TV | television |
TVA | Tennessee Valley Authority |
UH | upper half |
Ukr | Ukrainian |
Umbr | Umbrian |
UN | United Nations |
UNESCO | United Nations' Educational, Scientific and Cultural Organization |
univ | university |
US | United States |
USA | United States of America |
USN | United Stated Navy |
USNR | United States Naval Reserve |
USS | United States Ship |
USSR | Union of Soviet Socialist Republics |
usu | usual, usually |
v | velocity, verb, versus, vowel |
Va | Virginia |
var | variant |
vb | verb |
VF | vicar forane |
VG | vicar-general |
vi | verb intransitive |
v imper | verb imperative |
viz | videlicet |
VL | Vulgar Latin |
voc | vocative |
vocab | vocabulary |
vol | volume |
vt | verb transitive |
Vt | Vermont |
VT | variable time |
W | Welsh, west, western |
Wall | Walloon |
Wash | Washington |
Wed | Wednesday |
WFris | West Frisian |
WGmc | West Germanic |
Wis or Wisc | Wisconsin |
Wisd | Wisdom |
Wisd Sol | Wisdom of Solomon |
W. Va. | West Virginia |
Wyo | Wyoming |
YMCA | Young Men's Christian Association |
yr | year |
yrbk | yearbook |
YWCA | Young Women's Christian Association |
Zach | Zacharias |
Zech | Zechariah |
Zeph | Zephanian |
zool | zoological, zoology |