Possible phonemes

The theory of features answers the question of possible phonemes, saying that the segments which can be constructed using these features are all and the only possible phonemes. This gives a mathematical upper limit of 2n segments, given n binary features, so if there are twenty features (a reasonable number), there are 1,048,576 logically possible feature specifications, and this is quite a lot of segments. It also has to be physically possible to realize a segment, so the number of possible segments is smaller than this. Many segments can be imagined which are phonetically uninterpretable, such as one which is [+high, +low]. Such a segment is physically impossible since the tongue cannot be contradictorily raised and lowered at the same time, so the nonexistence of a large class of such segments is independently explained. Similarly, no segment can be [+cons, -high, -back, -ant, -cor]. A segment which is [+cons] is not a vowel or glide. The feature [-back] tells us that the segment would have a place of articulation in front of the velar position. [-ant] tells us that it must have a place of articulation behind the alveolar ridge, and [-high] tells us that it cannot be a palatal. Everything about this description suggests the vowel [e], except that it is [+consonantal], whereas vowels are [-consonantal]. No major constriction can be formed with the tongue in the position of [e]: hence this combination of features happens to be physically impossible. To be attested in a language, a segment must be both combinatorially possible, i.e. it must use just the features given by the theory, and physically possible.

Although the set of attested phonemes in human languages is quite large, there are significant limitations on what phonemes are possible. Retroflex consonants have the features [-anterior, +coronal, -distributed]. Recall the question whether a language could contrast two kinds of retroflex consonants, such as apical and subliminal retroflex as found in Hindi versus Telugu. According to this theory of features, such a contrast is impossible, since no feature is available to describe such a difference within a language. Phonetic differences across languages are possible because phonetic interpretation is not subject to the limitations of phonological feature theory. Were we to discover such a contrast, the theory of features would be challenged, because it has no mechanism for expressing such a distinction. Similarly, the differences attested in the phonetics of [u] and [ʊ] across languages are never found within a language. In a single language, the maximal contrast is between two such vowels, governed by the feature tense (or ATR). The fact that such differences exist at the phonetic level between languages, but are never exploited within a single language as a way to distinguish words, is an example of the difference between phonetic and phonological properties.

Thus one of the main goals of distinctive feature theory is providing a predictive framework for saying what contrasts will and will not be found in the phoneme systems of human languages.