Writings > Lexical Leavings


Mark Spahn writes:

The glottal-stop interjections uh-oh, uh-huh, and uh-uh

According to Webster's New World College Dictionary, Fourth Edition (2002), a glottal stop is defined as "a speech sound (IPA symbol [']) articulated by a momentary complete closing of the glottis [the opening between the vocal cords and the larynx]; it is sometimes heard as a variant for medial t (as in bottle or water) in some English dialects [in Liverpool, for example], and is the medial sound in the negative expression conventionally spelled unh-unh in English".

This definition is identical to the definition in the Third Edition (1988), except that in the earlier edition the IPA symbol is written not as a tiny apostrophe but as the top half of a question mark (i.e., ? without the dot at the bottom).

The above explanation of the pronunciation of unh-unh (meaning "no" and more commonly spelled uh-uh) is misleading: a glottal stop occurs not just in the middle of the word ("medial") but also at its beginning. This word consists of two syllables, each of which consists of a glottal stop followed by a short-u vowel (as in "up") that is nasalized (i.e., pronounced with part of the breath passing through the nose). Let's denote this syllable by /?uN/.  According to WNWCD4, the stresses on the two syllables are either /?uN"?uN"/ or /?uN"?uN'/, where the double-quote /"/ represents a primary stress, and the single-quote /'/ represents a secondary stress.

The opposite of uh-uh is the interjection uh-huh, which means "yes" and is pronounced /?uhu'/ or /?uhuN"/.

Another interjection whose pronunciation begins with a glottal stop (and has another glottal stop in the middle) is /uh-oh/, which is "used to signify sudden awareness of a problem or error and the resulting worry, alarm, etc."  It is typically followed by a remark like "Now we're in trouble." (A good Japanese equivalent for uh-oh is yabai.) This interjection uh-oh is pronounced /?u"?o'/.

The pronunciations given for these three interjections /uh-oh/, /uh-huh/, and /uh-uh/ in WNWCD4, in the American Heritage Dictionary (AHD), and in the Random House Unabridged Dictionary, Second Edition (RHUD2) are summarized as follows.
          WNWCD4           AHD               RHUD2   
uh-oh    /?u'?o"/       /?u"?o'/       /?u"?o'/
uh-huh   /?uNhuN"/      /?uN'huN"/     /?uhu'/, /?uNhuN"/
uh-uh    /?uN"?uN"/     /?uN"?uN'/     /?uN"?uN"/, /?uN"?uN'/
(Note: The WNWCD4 pronunciation for uh-oh seems to be a misprint; the primary stress should be on the first syllable.)

But the glottal stop marks ? that appear in the above table have been inserted; they do not appear in the pronunciations given in any of the above dictionaries.  This is surprising, because the glottal stops are essential to the pronunciation of these words.  (Just try pronouncing them without their glottal stops and see how weird and ununderstandable these words become.)

The pitch intonation of these interjections is so distinctive that they can be pronounced perfectly understandably while keeping the lips closed. Such a closed-mouth pronunciation is indicated by a spelling like "mm-hm" instead of "uh-huh".

Musical pitch seems to be more important with these interjections than with most words.  Here's an experiment we might perform: Ask music-savvy informants what notes (frequencies) they use in pronouncing the two syllables of uh-oh. That is, if they were duplicating the pronunciation of uh-oh on a piano (to convey the alarm of a cartoon character who has run off the edge of a cliff and belatedly realizes that there is no ground beneath him), what keys on the piano keyboard would be pressed? How would the two notes of uh-oh be denoted in musical notation? Among such music-savvy informants, surely the first note would be at a higher pitch than the second note, probably the musical interval (frequency ratio) between the two notes would be the same, and possibly the very notes themselves (e.g., E-flat above middle C, then middle C) would be the same for all informants.

I remember a brief article in Harper's magazine a year or two ago reporting on a phenomenon observed in college basketball arenas. When a player tries to throw the ball through the basket but misses the basket entirely, not even hitting the rim, he is said to make an "air ball". Upon seeing such a spectacularly bad play, the spectators regularly shout out, in unison, the jeer "Air ball!" What is remarkable is that the spectators are able to synchronize their shouted-out jeer with each other so accurately, and even synchronize the pitches of the two syllables (with some variation in the exact musical notes from one college arena to another).

(November 4, 2003)