A number of short sound recordings of spoken Sinampaiton are available. Samples and links for the examples used in this page will be added later.
Sinampaiton is an invented language, and is spoken (very poorly) by its creator alone (the author of this Web site). More information about Sinampaiton is available, though much of it is confused or out of date -- unfortunately, Sinampaiton seems to change from month to month, according to the whim of its sole speaker.
The field of phonology is (in a general sense) concerned with the sounds that the speakers of languages use: what they are and how they are put together to form utterances (sentences, sentence fragments, etc.). A full phonological description of a language might touch on subjects such as stress, pitch, and meter.
Phonetics, on the other hand, is more concerned with the sounds themselves and the ways in which they are produced: the physiology of the human speech apparatus (tongue, lips, teeth, etc.), the acoustical properties of sounds, and so on.
Obviously, phonology and phonetics are closely related. A study of phonology is ultimately meaningless without the physical connection to reality that phonetics provides. Because of this, a full phonological description of a language must necessarily include some aspects of phonetics.
The basic units of phonetics and phonology are phones (the actual sounds produced in speaking a language) and phonemes (the somewhat abstract sound units that underlie phones).
(Please be careful not to confuse (tele)phones1 and the phones2 of phonetics. Phones1 are often used to transmit sounds in the form of phones2, of which the English word phone1 and 2 is comprised!)
Sinampaiton has a fairly simple array of vowels (with a number of diphthongs) and a wide variety of consonants. The following features are particularly noteworthy:
Length contrast in the vowels: a, aa; e, ee; etc.
Aspiration contrast in the voiceless stops: p, ph; t, th; k, kh
Prenasalized consonants: mp, mb; nt, nd, etc. (including a prenasalized lateral approximant nl)
A rich set of fricatives and affricates: s, ts, dz; sh, ch, j, etc. (including a voiced lateral fricative ll)
A phonological inventory is a catalog of the phonemes of a language, usually arranged (as here) in tabular form according to the dimensions of length, place of articulation within the mouth (actually, the human speech apparatus in general: tongue, lips, etc.).
These are the "simple" vowels. Their sound qualities are very similar to those in Spanish. The long vowels are simply longer versions of the short vowels; no difference in quality is discernible (or at least not to my ears).
These are the "complex" vowels. In a falling diphthong, the second vowel in the pair is a semivowel; in a rising diphthong, the first vowel is a semivowel.
The diphthong ui may be difficult for English speakers. Its pronunciation is similar to the vowels in the English word gooey, but the u is much shorter than the oo in English.
2 Voiced alveolar lateral fricative (the same sound as Zulu dl: like Welsh ll, but voiced).
3 An uncommon and rather elusive phoneme, occurring in some dialects as nd, in others as ndz, and occasionally as nll (prenasalised voiced alveolar lateral affricate).
4 Like the r is Spanish pero, not trilled like the rr in perro.
In Sinampaiton, a light stress accent is given to the first syllable in so-called content words (nouns and verbs and certain other polysyllabic words). These stress accents have a pitch component, too, but this hasn't been worked out. My guess at this point is that even pitch is used for the topic of a sentence (what the sentence is about) and raised pitch for its focus (new material brought into the discourse), if it has one. Other details will doubtless emerge in time.
A single phoneme may have more than one phonetic realization or phone; these phones are called allophones of the phoneme. The distinction between phones and phonemes is an important one, so it must be possible to show this distinction in writing. The usual convention is to use slashes to denote a phoneme (or sequence of phonemes) and brackets to denote a phone (or sequence of phones). Similarly, a phonemic transcription of a sentence uses phonemes to represent the phonological nature of the word, while a phonetic transcription uses phones.) The following example illustrates this usage.
The English phoneme /p/ is sometimes aspirated (accompanied by a following breath of air) and sometimes unaspirated. In other words, English /p/ has two allophones: [p] and [ph]. (The superscripted h indicates aspiration of the preceding phoneme.) You can hear the difference by comparing the pronunciations of the English words pit and spit -- in the former, the p is aspirated, while in the later, it isn't. By putting your hand right in front of your mouth while you pronounce these words, you can even feel the difference.
In some cases, two or more distinct phonemes may be realized as a single phone under some conditions and as distinct phones under other conditions. For example, in many dialects of English the phonemes /t/ and /d/ when found between vowels are both realized as an alveolar flap (the sound of untrilled r in Spanish). Under different conditions (as at the beginning of a word) /t/ and /d/ are realized as distinct phones. Compare the pronunciations of the English words cedar and seater (as in two seater), in which /t/ and /d/ sound the same; then compare those of tin and din, in which /t/ and /d/ are quite different.
When a language changes so that two different phonemes share the same phone under all conditions (between vowels, after a voiced consonant, or whatever), then the phonemes have merged to form a single phoneme.
Phonotactics is the study or description of the ways in which the phonemes of a language are combined. Different languages have different constraints on what consonants can occur next to each other, or at the beginning or end of a word. For example, in Japanese the only consonant that may be followed by another consonant is n (usually transliterated as m before p or b), and not all consonants may follow n. Russian, on the other hand, has far fewer constraints on consonant clusters.
Very little clustering of consonants is allowed in Sinampaiton, unless one considers prenasalised consonants to be nasal + consonant clusters. (In fact, I see medial nt, etc. as n + nt, etc., though I'm not sure that's the best analysis.).
In most varieties of the language, the only clusters that occur are ps, ks, pr, and kr. These are quite uncommon, being foreign in origin for the most part; they're described in more detail below. Some dialects don't have quite the same distinction of aspirated vs. unaspirated stops, and some have an additional series of prenasalized stops. Unfortunately, I haven't worked this all out yet.
The basic phonological structure of a word in Sinampaiton is fairly restrictive; a word must end in a vowel or n. A non-final syllable must end in a vowel. (Or, depending on how you analyse prenasalized consonants between vowels, in a homorganic nasal N (spelled m or n, depending on the following consonant). For example, kambe may be analyzed either as the syllables ka + mbe or kaN + mbe. Which analysis is more correct depends in all likelihood on characteristics of prosody in Sinampaiton, which I haven't worked on at all.)
The basic formula for a word in Sinampaiton is as follows (using a regular expression-like notation). This formula can be made to work with either type of analysis of prenasalized consonants between vowels.
word = C?V(N?CV)*n?
Here, C represents any consonant (or consonant cluster), V represents a vowel, and N represents a hoorganic nasal. Parentheses are used to group elements of the word; ? means that the preceding element may or may not occur, while * means that the preceding element may occur once, more than once, or not at all.
The following examples may be illustrative:
|au||V||A diphthong or long vowel functions as a single V|
|mpin||CVn||A prenasalised phoneme functions as a single C|
|kambo||CVNCV||If analysed as kaN + mbo|
There are no vowel clusters within a Sinampaiton word.
The clusters ps and ks are sometimes replaced by ts, which is a phoneme in its own right and not a cluster. They are sometimes heard as p+sh and k+sh, or replaced by the affricate ch.
The clusters pr and kr are often replaced by ph and kh. When they aren't, p and k are generally aspirated, while r is devoiced and trilled. In Evan Kirshenbaum's IPA/ASCII schema (see the sci.lang FAQ) these would be represented as p<h>r<trl><vls> and k<h>r<trl><vls>. In some dialects, the r even becomes uvular. It may be better to consider pr and kr as phonemes in their own right in these dialects.
It is perilous (though often convenient) to pretend that words occur in isolation. A phonological word can't be fully described without mention of its environment. In many languages, the ends and beginnings of words interact in strange ways. The term sandhi is taken from Sanskrit, which is just such a language. The details of this aspect of Sinampaiton phonology haven't been answered yet in any detail, though a few examples will lead one to the pertinent questions:
|Milan ku su suin se bute
'Here's a red book'
|What happens when consonants (here, k and s) follow a word-final n?|
|Thalonge dzuna ogomun
'I want to write (something)'
|What happens when vowels (here, a and o) collide?|
Last modified Saturday, April 24, 2004 at 16:40:54 GMT -0500