 Dialectal change can be determined by the linguistic analysis of regional vowel systems. Why not consonants, you might ask? Well, consonantal variation does not involve variation in terms of degree, but primarily follows a binary principle. That is, either speakers use a particular consonant or consonantal feature, or they don't. Look at these examples here. Now, depending on the dialect you speak, the English dialect you speak, you might realize orthographic are either as an alveolar approximate, as an RP, or as an alveolar flap, as for example in Scottish English, or take the German orthographic CH, especially at the beginning of words. Now, you might in standard German use the palatal fricative as in China, China chemistry, whereas in some selected variants of German, we might use the Vilaplosive as in Kina or Kimi. So, we have the choice between two variants in each case, an either or but nothing in between decision. Now, vocali variation by contrast is gradual. Take the two low back vowels as presented here, or as in British English, caught versus or as in British English caught. Now, there is a clear cut difference between them in RP. However, there are varieties where they have merged to become just one phoneme. In many parts of North America, for example, in Canada or the west of the United States. However, there are in-betweens. For example, in some parts of the United States where there is still a difference, even if this difference is not as big as in RP. So, vowels may involve gradual differences. That is, a vowel can be more or less front, can be more or less high or low. And this allows us to build phonological maps, not only in an either or fashion, but in terms of gradiences. However, gradual variation is extremely difficult to identify. Let's nevertheless try and do it. Here you have three North American vowels. Let's listen first. Bra, cut, caught. Now, as well-educated linguists, you must be in a position to locate them on a vowel chart. So, let's move them into their position. So, I would say that this is a back vowel, which is slightly fronted. Cut. This is a low vowel, which is lower than cardinal vowel number six. Cut. And the one in caught is a vowel, which is close to cardinal vowel number six. Such impressionistic judgments are relatively unreliable. Vocalic studies are thus mostly supported by precise spectrographic analysis of the vocalic form and frequencies involved. One question in comparative phonology concerns the general organization of vocalic systems. More precisely, the predictability of vocali changes. That is, are there tendencies that vowels arrange themselves according to particular patterns? Well, let's look at this problem in more detail. Most languages have their vowels evenly and widely distributed over the chart, exhibiting a triangular vowel pattern. So, this would be the simplest one. We have one front vowel, one back vowel, which corresponds to this front vowel. So, it is produced with the same tongue height. And we have one central vowel like this. So, this would then constitute a neat triangular pattern. Such a pattern can be found in a language such as Green Landing. However, such a pattern is the exception rather than the rule. I will now present two vocalic patterns, one consisting of five vowels and one consisting of seven vowels. And the languages are taken from the virtual linguistics campus language index, namely Spanish with five vowels and Romanian with seven vowels. Now here, we have Spanish with five vowels. Vida, peso, paso, vota, puro. And again, we have a nice triangular pattern with a central vowel at the bottom and two front vowels which have a back vowel counterpart. Let's look at Romanian next. Now this speaker from Sibiu in Romania has seven vowels. Inima, bere, cap, ok, unde. This would be almost identical to Spanish, but now we have two further vowels, namely two central vowels. Tintec, vaje. So again, we have our triangular pattern. But this time we also have in the middle vowels that can be added to the pattern. Romanian belongs to the majority of languages that have between five and seven vowels. These examples show that languages seem to develop geometrical vowel patterns. Each front vowel has a corresponding back vowel and vice versa. This phenomenon has become known as pattern congruity. But what about those languages where this is superficially not the case? For example, RP. Let's look at the tense vowels in RP. Now the tense vowels in RP are the green ones, the light green ones. Well here, we have of course a front vowel with a back vowel counterpart. But what about this one? Where is the counterpart? And furthermore, this back vowel, which for many speakers is even further back than for this one, what about the counterpart? Well, we might argue that this counterpart is developing right now. For example, if you look at some varieties of English, North American, for example, then you know that the diphthong A in British English, as in say, treated as a monothong in North American English, and that the lax low front vowel, as in ah, as in ash, is raised in some varieties of English and is becoming longer as in man. So the congruent pattern seems to re-emerge. Or let's look at a case in German now. The German tense monothongs constitute an almost perfect pattern. Let's look at it. Now here we have a speaker from Central Germany speaking standard German. Let's listen to the two vowels I have in focus first. Beat. Now this means the bed in the garden. It's clearly a high front vowel. Beat. And this. Welle. Means English select. Welle. It's a mid vowel. But what about the counterparts? The back vowel counterparts? Well, this one can clearly be associated with this one. But what about the counterpart here? It doesn't exist. And for this reason there's a sound shift going on in standard German where many Germans merge these two vowels. So they become identical just like the two low back vowels in North American English. Let's listen to one example. Now here we have a speaker from Berlin. Well, and this speaker has clearly merged these two vowels. Beat. Heeler. And now the pattern is in line again because we do not have to take care of these mid low vowels anymore. Having discussed the distribution of vowels on the cardinal vowel chart, let us now look at the mechanisms of vocali change. Now in many languages, vocali changes often affect the entire vowel system and not just individual vowels. In North American English, for example, the status of the short o as in cot and dog and the short a as in ash and hat will influence the behavior of the rest of the vowel system. This system effect can be observed in the so-called Northern Cities Shift, NCS, which is a series of innovations in the vowels of the English spoken in the urban centers around the Great Lakes. For example, Detroit, Chicago, and so on. The result of the Northern Cities Shift is that each of a series of vowels receives a new place of articulation along the parameters of tongue advancement, front-back, or tongue height. The low vowels are advanced. So here are the low vowels. They move to the front. Well, and the high vowels and the mid vowels, they are retracted. So here. The most well-known system effect in the history of English is the Great Vowel Shift, GVS. Roughly between 1350 and 1555 of the seven long monophthongs of middle English were raised and two, two top ones here, they were diphthongized. System effects are triggered by individual vocaliate changes. So let's finally look at these in more detail. The first group of vocaliate changes that I want to introduce involves vowels where the change has no particular direction or location on the chart. This one here, this picture, stands for the effect of merging two vowels into one. Well, the low-back merger, as in cotton caught, or the e-a-merger in the south of the United States where pin and pen are pronounced identically may be examples. The opposite is called splitting, or split. And a typical example of this is the short a-split that affects the English spoken in New York City. In the southern states of the United States, we have an effect that can be called breaking where one vowel, especially the vowel of short a, is a little bit raised and then it is broken into several parts with a glide in between. This effect has become known as the southern drawl and it affects words like can't which would be pronounced as can't in the south of the United States. In addition to these non-positional changes, we have vowel shifts that can be located on the chart. For example, we might have something we want to call raising. So if a vowel changes from a low-tongue position to a higher-tongue position, it is raised. Now a typical effect would be the short a-raising where something like man becomes man. The opposite would be lowering. And then finally, if vowels move to the center, well, then we have an effect of centering. Let's summarize. Vowels systems seem to organize themselves in a systematic way. Congrant patterns seem to be the target of this self-organization process. And as soon as incongrant patterns emerge through general effects of language change, sound shifts come into action that repair these deficits by re-establishing a congruent pattern. These sound shifts may affect just two vowels or the entire vocalic system. With the great vowel shift and the northern city shift, we have presented some well-known examples from the history of English and from present day English.