Hi. So far, we have concentrated on phonetics. We have looked at the precise way in which consonants are articulated. Let's now move to phonology. Let's try to see how such distinctions also play a role in language. These distinctions we've seen, such as place of articulation, manner of articulation, and voicing, in phonology are usually called features. Linguists assume that sounds are constituted of such features, so that in a language, what t really is is just a combination of being alveolar, being plosive, and being voiceless. The true atoms of language, the smallest things, are just those features. In this video, we're going to look a little bit about what the implications of this assumption are. Well, one kind of result comes from phonological activity, as it's called. In languages, sometimes one sound changes into another sound. This can then usually be described as one feature changes into another feature. Look at the following example which I took from Turkish. What you see here is various words in two different cases, a nominative, that's the NOM, and a dative, that's the DAT case. You see that this DAT case has an ending, an a. But you also see that something changes in the last consonant of the stem. For instance, in Kalip or Kalib-a, in Kanat or Kanad-a. There's a t versus a d, or a p versus a b. What kind of change is that? Well, remember from my discussion of articulation in the second video that the difference between T and D Is one of voicing, of whether or not my vocal cords vibrate. And that's exactly the difference we find all the time. At the end of a word in Turkish, every consonant is always voiceless. You never vibrate your vocal cords. Within a word, you might. So there is a change that's the phonological activity going on and it's a change from being voiceless to being voiced. There's another way in which features or these dimensions of speech play a role in human language, and that we can find out if we study tables of consonants, such as the one we saw for English in the first video of this module. If you study such a table closely, you will very often see that it's quite symmetric, that the consonants line up in nice rows and columns. That's true for English, but it's also true for languages which have slightly fewer consonants, like Hawaiian. Here's the table of consonants for Hawaiian. If you look at this table, you see various things. First you see that the language really doesn't have many consonants. You also see that there is no difference in the language between the t and the k sound. If you listen to Hawaiian very closely, you will hear both sounds, but there's not going to be a difference. Sometimes people say ta, sometimes they say ka, and they mean the same thing. Features also help us to describe and understand the path of acquisition, the order in which children learn the sounds of their language. A common path for children learning English, for instance, is that first they know the sounds p, t. What that means is they have then only learned the features voiceless, plosive, labial and coronal. Think about this for a little bit. If you have those features, you can make two sounds, p and t. What can happen after this is that children learn the feature nasal. When they have done so, they can combine this one feature with the features they already have. They have p and t, they can combine it with nasal, then they can make two sounds, m and n. That's usually exactly what happens. They double their inventory. Suppose then they learn how to make fricatives. Well, they can turn everything into fricatives at the same time. So they will learn to say f and s as well. They don't learn sounds one at a time, they learn places of articulation and manners of articulation one at a time. Features play a role also in speech errors when people make mistakes and that's what we do all the time. In many speech errors, the sounds of language get interchanged. Sometimes this is kind of uninteresting because it just involves whole sounds. Somebody's trying to say brake fluid and says blake fruid instead. What happens is the r and the l get interchanged here. But sometimes it also involves features rather than complete sounds. Somebody's trying to say Cedars of Lebanon, but says Cedars of Lemadon instead. We have a non-nasal b and then a nasal n, but it changes into a nasal m and then a non-nasal d. So it's as if the feature nasal moves from the n to the b. Another example is I knew a child who always said skabetti instead of spaghetti. In spaghetti, the p is voiceless and the g is voiced, but in skabetti, as she said it, the k is voiceless and the b is voiced. So it's, in this case, as if the feature voice has moved from one sound to the next. In this video, we have seen that place of articulation, manner of articulation, voicing are not just nice ways to describe the way we pronounce sounds, but they actually play a role in language as the atoms of language, which we call features. I've given you some evidence for this from phonological activity, as in our Turkish example, from the asymmetry of consonants, like in Hawaiian, from the passive acquisition which children may take, and from speech errors. In the next video, we will discuss sounds a little bit more with Marten and Inge, but from an interesting new perspective, namely, the perspective of the written word.