Possible differences in sounds

One way to approach the question is to collect samples of the sounds of all of the languages in the world. This search (which has never been conducted) would reveal massive repetition, and would probably reveal that the segment [m] in English is exactly the same as the segment [m] in French, German, Tübatülabal, Arabic, Swahili, Chinese, and innumerable other languages. It would also reveal differences, some of them perhaps a bit surprising. Given the richness of our transcriptional resources for notating phonetic differences between segments, you might expect that if a collection of languages had the same vowels transcribed as [i] and [ɪ], then these vowels should sound the same. This is not so.

Varieties of phonetic [i] vs. [ɪ]. Many languages have this pair of vowels; for example, Matuumbi has [i] and [ɪ]. But the actual pronunciation of [i] vs. [ɪ] differs between English and Matuumbi. Matuumbi [i] is higher than in English, and Matuumbi [ɪ] is a bit lower than English [ɪ] – to some people it almost sounds like [e] (but is clearly different from [e], even the “pure” [e] found in Spanish). This might force us to introduce new symbols, so that we can accurately represent these distinctions. (This is done in publications on Matuumbi, where the difference is notated as “extreme” i̧, u̧ versus “regular” i, u.) Before we embark on a program of adding new symbols, we should be sure that we know how many symbols to add. It turns out that the pronunciation of [i] and [ɪ] differs in many languages: these vowels exist in English, Kamba, Lomwe, Matuumbi, Bari, Kipsigis, Didinga, and Sotho, and their actual pronunciation differs in each language.

You do not have to go very far into exotic languages to find this phonetic difference, for the difference between English [i] and German [i] is also very noticeable, and is something that a language learner must master to develop a good German or English accent. Although the differences may be difficult for the untrained ear to perceive at first, they are consistent, physically measurable, and reproducible by speakers. If written symbols are to represent phonetic differences between languages, a totally accurate transcription should represent these differences. To represent just this range of vowel differences involving [i] and [ɪ], over a dozen new symbols would need to be introduced. Yet we do not introduce large numbers of new symbols to express these differences in pronunciations, because phonological symbols do not represent the precise phonetic properties of the sounds in a language, they only represent the essential contrast between sounds

Other variants of sounds. Similar variation exists with other phonetic categories. The retroflex consonants of Telugu, Hindi, and Koti are all pronounced differently. Hindi has what might be called “mild” retroflexion, where the tip of the tongue is placed just behind the alveolar ridge, while in Telugu, the tip of the tongue is further back and contact is made between the palate and the underside of the tongue (sublaminal); in Koti, the tongue is placed further forward, but is also sublaminal. Finnish, Norwegian, and English contrast the vowels [a] and [æ], but in each of these languages the vowels are pronounced in a slightly different way. The voiced velar fricative [γ] found in Arabic, Spanish, and the Kurdish language Hawrami are all phonetically different in subtle but audible ways.

The important details of speech. Although languages can differ substantially in the details of how their sounds are pronounced, there are limits on the types of sound differences which can be exploited contrastively, i.e. can form the basis for making differences in meaning. Language can contrast tense [i] and lax [ɪ], but cannot further contrast a hyper-tense high vowel (like that found in Matuumbi), which we might write as [i+ ], with plain tense [i] as in English, or hyper-lax [ɪ - ] as in Matuumbi with plain lax [ɪ] as found in English. Within a language, you find at most [i] vs. [ɪ]. Languages can have one series of retroflex consonants, and cannot contrast Hindi-style [ʈ] with a Telugu-style phoneme which we might notate as [ʈ + ]. The phonology simply has “retroflex,” and it is up to the phonetic component of a language to say exactly how a retroflex consonant is pronounced.

It is important to emphasize that such phonetic details are not too subtle to hear. The difference between various types of retroflex consonants is quite audible – otherwise, people could not learn the typical pronunciation of retroflex consonants in their language – and the difference between English and German [i] is appreciable. Children learning German can hear and reproduce German [i] accurately. Speakers can also tell when someone mispronounces a German [i] as an English [i], and bilingual German– English speakers can easily switch between the two phonetic vowels.

One thing that phonological theory wants to know is: what is a possible phoneme? How might we answer this? We could look at all languages and publish a list. A monumental difficulty with that is that there are nearly 7,000 languages, but useful information on around only 10 percent of these languages. Worse, this could only say what phonemic contrasts happen to exist at the present. A scientific account of language does not just ask what has been actually observed, it asks about the fundamental nature of language, including potential sounds which may have existed in a language spoken 1,000 years ago, or some future language which will be spoken 1,000 years hence. We are not just interested in observation, we are interested in prediction.

In this connection, consider whether a “bilabial click” is a possible phoneme. We symbolize it as [ʘ] – it is like a kiss, but with the lips flat as for [m], not protruded as for [w]. Virtually all languages have bilabial consonants, and we know of dozens of languages with click consonants (Dahalo, Sotho, Zulu, Xhosa, Khoekhoe), so the question is whether the combination of concepts “bilabial” and “click” can define a phoneme. As it happens, we know that such a sound does exist, but only in two closely related languages, !Xoo and Eastern 6¼Hoan, members of the Khoisan language family. These languages have under 5,000 speakers combined, and given socioeconomic factors where these languages are spoken (Namibia and Botswana), it is likely that the languages will no longer be spoken in 200 years. We are fortunate in this case that we have information on these languages which allows us to say that this is a phoneme, but things could have turned out differently. The languages could easily have died out without having been recorded, and then we would wrongly conclude that a bilabial click is not a possible phoneme because it has not been observed. We need a principled, theoretical basis for saying what we think might be observed.

Predictions versus observations. A list of facts is scientifically uninteresting. A basic goal of science is to have knowledge that goes beyond what has been observed, because we believe that the universe obeys general laws. A list might be helpful in building a theory, but we would not want to stop with a list, because it would give us no explanation why that particular list, as opposed to some other arbitrary list, should constitute the possible phonemes of language. The question “what is a possible phoneme?” should thus be answered by reference to a general theory of what speech sounds are made of, just as a theory of “possible atoms” is based on a general theory of what makes up atoms and rules for putting those bits together. Science is not simply the accumulation and sorting of facts, but rather the attempt to discover laws that regulate the universe. Such laws make predictions about things that we have yet to observe: certain things should be found, other things should never be found.

The Law of Gravity predicts that a rock will fall to earth, which says what it will do and by implication what it will not do: it also won’t go up or sideways. Physicists have observed that subatomic particles decay into other particles. Particles have an electrical charge – positive, negative or neutral – and there is a physical law that the charge of a particle is preserved when it decays (adding up the charges of the decay products).

The particle known as a “kaon” (K) can be positive (K+ ), negative (K- ) or neutral (K0 ); a kaon can decay into other particles known as “pions” (π) which also can be positive (π+ ), negative (π- ) or neutral (π0 ). Thus a neutral kaon may become a positive pion and a negative pion (K0 ! π+ + π- ) or it may become one positive, one negative, and one neutral pion (K0 ! π+ + π- + π0 ), because in both cases the positives and negatives cancel out and the sum of charges is neutral (0). The Law of Conservation of Charge allows these patterns of decay, and prohibits a neutral kaon from becoming two positive pions (K0 ! π+ + π+ ). In the myriad cases of particle decay which have been observed experimentally, none violates this law which predicts what can happen and what cannot.

Analogously, phonological theory seeks to discover the laws for building phonemes, which predict what phonemes can be found in languages. We will see that theory, after considering a related question which defines phonology.