Posted: Apr 08, 2012 9:32 pm
PHONETICS AND PHONOLOGY
1. Phonetics, a very bare-bones introduction part I.
Phonetics is basically "how speech sounds come about and their properties".
The majority of languages in use consist of strings of sound. We produce these sounds by use of our lungs, oral and nasal passages, the tongue, lips and uvula, and so on.
The main method of generating sounds is by expelling air from the lungs, which by Bernoulli's law creates vibration whenever it runs into some kind of constriction. Our vocal cords is the source of what we call voicing, a quality that is shared by consonants such as l, w, d, n, m and almost every vowel uttered in a normal voice in most languages around the world. The sounds thus produced are fairly regular - they're tones. This contrasts with less regular sounds, e.g. sounds produced by friction. Some sounds exist in various languages that do not use pulmonary exhalation - clicks are one such type of sounds.
Basically, the mouth and nose are two tubes along which air passes, sometimes carrying a tone. Turns out the musculature we have can do pretty cool things, such as modulate the timbre, or obstruct the passage of air so that different noises are produced. We can open the tube up wider, and get our vowels more towards an "aw" or "aaargh" or "aaaa!"-like quality, or contract the tube and get something like "woo" or "you" or "eee". The difference between woo and yeeee is in how far towards the back (wooo) or the front (yeeee) of vowel space the narrowest bits are. Other tricks exist, such as lengthening the front part of the tube by rounding your lips, (Swedish, Estonian, Finnish and German ö, German and Estonian ü, Finnish and Swedish y, French u, Danish and Norwegian ø, ...). For some reason, most languages in the world prefer to round their back vowels (o, u), yet not round their front vowels (i,e). A minority have an added set of front rounded vowels (ü,ö,...), and some have an unrounded back vowel (Japanese 'u' apparently is such a vowel).
Another thing we can do is create a bit of double resonance by diverting some of the flow and tone into the nasal cavity, getting, you know, those French-style nasal vowels. (Also present in Portuguese and any number of other languages; ultimately, careful listening would reveal that they also are present in words like "nun", "none", etc in English.)
We can also stop the air flow temporarily - but keep on exhaling, letting a bit of pressure build up - and then release the pressure. We get what's called a stop. The sounds p, t, k, b, d, and g (in their most typical forms) are all results of this. The main thing that distinguishes them is where in the mouth they happen - the shape of the tube behind (and in part, in front of) the occlusion. In English, there's four locations where this occlusion occurs. Between the lips, at the alveolar ridge (roughly), at the soft palate/velum, and at the vocal cords. The last one is exceptional, in that since the vocal cords normally produce the tone associated with voicing, stopping the air stream at that point makes it impossible to have voicing going on simultaneously. Now, this only really distinguishes four consonants in English: p and b are pronounced with the mouth in the same position, d and t likewise and g and k likewise. (Don't mind the fact that the letter g sometimes marks other sounds, letters are a secondary thing, and will be dealt with later), the glottal one is usually written using the letter <t>, and its properties are better left for later.
These three pairs are distinguished by whether the vocal cords vibrate while they're produced, and by how long it takes from the release of the pressure until the vibrations resume. Not all languages distinguish k vs. g, b vs. p based on whether the vocal cords vibrate. Some distinguish them depending on how long it takes for the voicing of the vowel to resume. Some just have p t k, and some of those may have b d and g as functionally equivalent variations of p t and k - e.g. apa and aba would be the same word, and a speaker of those languages would perceive the different sounds in the middle as the same sound. Likewise, sounds we don't distinguish are considered distinct sounds in other languages. Russian, for instance, pretty much doubles the number of consonants, simply by distinguishing whether the tongue is raised towards the palate or not during pronunciation. (Although this also creates a bit of an off-glide effect in the next vowel. As well as having other effects on the vowels as well.)
Another kind of sound we can get when we stop the airflow through the oral cavity entirely is what we get when we redirect it through the nasal cavity: m, n and ŋ (written as ng in English, usually). These are therefore also called nasal stops. They're normally voiced in almost all languages that have them (which is pretty close to every language, although some don't have all three, and some have more. There's no special thing about the three English has.)
There's something to say about where the obstruction goes. One obvious obstruction, of course, is obtained by closing your lips. This is one of the few obstructions everyone can do identically - for most of the other ones, exact position in the mouth probably varies a bit between repeated productions of the same sound even for one individual. t and k are less obvious ones, and their exact position varies a bit in different languages. In English, the exact position of k in the words kit and cool is somewhat different. Essentially, i is articulated a bit in front of where k usually goes, and this kind of pulls k forward, whereas oo doesn't much move the k from its usual spot, as it's quite close to where k usually goes.
Now, speaking of sounds like I did there - i, oo, k ... is unclear and confusing. The letter i has any number of pronunciations, and it's also unhelpful that these pronunciations can vary by region. I could of course say something utterly useless like "like the 'is' in how a Glaswegian would pronounce 'haggis'", and less than a handful would know what that sounds like, and they'd probably disagree with each other to boot. It's also quite useless since even if I know a sound and exactly how it is articulated, there's no guarantee I know that some dialect has it, and nothing guarantees that this is a dialect you could be expected to know about, etc.
The scholarly method of dealing with this is to describe it in terms of articulation - what organ touches what organ, what manner is the air stream modulated, etc. We divide the top of the mouth into segments - lip, tip of the teeth, backside of the teeth, alveolar ridge, hard palate, soft palate, uvula, (and then we get into the throat, which has a few parts too - only one that has any relevance in English phonology now, though - the glottis, although some of the other parts have some slight relevance to English phonetics). The lower part of the mouth is also divided - this is mostly really the tongue. Usually these are given as adjectives, as you'll usually find them in adjectival form when describing a sound. Since the tongue is relatively mobile, the idea is "a part of the tongue is moved towards some part of the top of the mouth", so essentially the tongue is the active part of the process of articulation.
Now we get to combine these to some extent: we get subapical alveolar, we get apical palatal, etc. Not all combinations are possible, though, because the tongue isn't infinitely flexible. The most flexible of the active articulators are the subapical, apical and laminal ones, usually termed the coronal ones. There's some extra orientations and whatnot available in the coronal area, but those are seldom spoken of, and they are not of very great interest.
As it's late right now, I'll resume this tomorrow, when I get more into manner of articulation as well as vowels - a trickier thing to describe.
1. Phonetics, a very bare-bones introduction part I.
Phonetics is basically "how speech sounds come about and their properties".
The majority of languages in use consist of strings of sound. We produce these sounds by use of our lungs, oral and nasal passages, the tongue, lips and uvula, and so on.
The main method of generating sounds is by expelling air from the lungs, which by Bernoulli's law creates vibration whenever it runs into some kind of constriction. Our vocal cords is the source of what we call voicing, a quality that is shared by consonants such as l, w, d, n, m and almost every vowel uttered in a normal voice in most languages around the world. The sounds thus produced are fairly regular - they're tones. This contrasts with less regular sounds, e.g. sounds produced by friction. Some sounds exist in various languages that do not use pulmonary exhalation - clicks are one such type of sounds.
Basically, the mouth and nose are two tubes along which air passes, sometimes carrying a tone. Turns out the musculature we have can do pretty cool things, such as modulate the timbre, or obstruct the passage of air so that different noises are produced. We can open the tube up wider, and get our vowels more towards an "aw" or "aaargh" or "aaaa!"-like quality, or contract the tube and get something like "woo" or "you" or "eee". The difference between woo and yeeee is in how far towards the back (wooo) or the front (yeeee) of vowel space the narrowest bits are. Other tricks exist, such as lengthening the front part of the tube by rounding your lips, (Swedish, Estonian, Finnish and German ö, German and Estonian ü, Finnish and Swedish y, French u, Danish and Norwegian ø, ...). For some reason, most languages in the world prefer to round their back vowels (o, u), yet not round their front vowels (i,e). A minority have an added set of front rounded vowels (ü,ö,...), and some have an unrounded back vowel (Japanese 'u' apparently is such a vowel).
Another thing we can do is create a bit of double resonance by diverting some of the flow and tone into the nasal cavity, getting, you know, those French-style nasal vowels. (Also present in Portuguese and any number of other languages; ultimately, careful listening would reveal that they also are present in words like "nun", "none", etc in English.)
We can also stop the air flow temporarily - but keep on exhaling, letting a bit of pressure build up - and then release the pressure. We get what's called a stop. The sounds p, t, k, b, d, and g (in their most typical forms) are all results of this. The main thing that distinguishes them is where in the mouth they happen - the shape of the tube behind (and in part, in front of) the occlusion. In English, there's four locations where this occlusion occurs. Between the lips, at the alveolar ridge (roughly), at the soft palate/velum, and at the vocal cords. The last one is exceptional, in that since the vocal cords normally produce the tone associated with voicing, stopping the air stream at that point makes it impossible to have voicing going on simultaneously. Now, this only really distinguishes four consonants in English: p and b are pronounced with the mouth in the same position, d and t likewise and g and k likewise. (Don't mind the fact that the letter g sometimes marks other sounds, letters are a secondary thing, and will be dealt with later), the glottal one is usually written using the letter <t>, and its properties are better left for later.
These three pairs are distinguished by whether the vocal cords vibrate while they're produced, and by how long it takes from the release of the pressure until the vibrations resume. Not all languages distinguish k vs. g, b vs. p based on whether the vocal cords vibrate. Some distinguish them depending on how long it takes for the voicing of the vowel to resume. Some just have p t k, and some of those may have b d and g as functionally equivalent variations of p t and k - e.g. apa and aba would be the same word, and a speaker of those languages would perceive the different sounds in the middle as the same sound. Likewise, sounds we don't distinguish are considered distinct sounds in other languages. Russian, for instance, pretty much doubles the number of consonants, simply by distinguishing whether the tongue is raised towards the palate or not during pronunciation. (Although this also creates a bit of an off-glide effect in the next vowel. As well as having other effects on the vowels as well.)
Another kind of sound we can get when we stop the airflow through the oral cavity entirely is what we get when we redirect it through the nasal cavity: m, n and ŋ (written as ng in English, usually). These are therefore also called nasal stops. They're normally voiced in almost all languages that have them (which is pretty close to every language, although some don't have all three, and some have more. There's no special thing about the three English has.)
There's something to say about where the obstruction goes. One obvious obstruction, of course, is obtained by closing your lips. This is one of the few obstructions everyone can do identically - for most of the other ones, exact position in the mouth probably varies a bit between repeated productions of the same sound even for one individual. t and k are less obvious ones, and their exact position varies a bit in different languages. In English, the exact position of k in the words kit and cool is somewhat different. Essentially, i is articulated a bit in front of where k usually goes, and this kind of pulls k forward, whereas oo doesn't much move the k from its usual spot, as it's quite close to where k usually goes.
Now, speaking of sounds like I did there - i, oo, k ... is unclear and confusing. The letter i has any number of pronunciations, and it's also unhelpful that these pronunciations can vary by region. I could of course say something utterly useless like "like the 'is' in how a Glaswegian would pronounce 'haggis'", and less than a handful would know what that sounds like, and they'd probably disagree with each other to boot. It's also quite useless since even if I know a sound and exactly how it is articulated, there's no guarantee I know that some dialect has it, and nothing guarantees that this is a dialect you could be expected to know about, etc.
The scholarly method of dealing with this is to describe it in terms of articulation - what organ touches what organ, what manner is the air stream modulated, etc. We divide the top of the mouth into segments - lip, tip of the teeth, backside of the teeth, alveolar ridge, hard palate, soft palate, uvula, (and then we get into the throat, which has a few parts too - only one that has any relevance in English phonology now, though - the glottis, although some of the other parts have some slight relevance to English phonetics). The lower part of the mouth is also divided - this is mostly really the tongue. Usually these are given as adjectives, as you'll usually find them in adjectival form when describing a sound. Since the tongue is relatively mobile, the idea is "a part of the tongue is moved towards some part of the top of the mouth", so essentially the tongue is the active part of the process of articulation.
- labial (pretty much exclusively though in the form "bilabial", meaning using both lips
- (interdental)
- subapical, using the bit of the tongue just under the tip
- apical. using the tip of the tongue
- laminal, using the blade of the tongue, the bid just behind the tip
- dorsal, using the middle part of the tongue
- radical, using the root of the tongue
Now we get to combine these to some extent: we get subapical alveolar, we get apical palatal, etc. Not all combinations are possible, though, because the tongue isn't infinitely flexible. The most flexible of the active articulators are the subapical, apical and laminal ones, usually termed the coronal ones. There's some extra orientations and whatnot available in the coronal area, but those are seldom spoken of, and they are not of very great interest.
As it's late right now, I'll resume this tomorrow, when I get more into manner of articulation as well as vowels - a trickier thing to describe.