 Yeah, thank you. About Nuristani, the only problem is I don't know very much about these group of languages yet. We are just in the process of initiating a research project about Nuristani languages at the University of Cologne. So I will talk about what we would like to know about the Nuristani languages, provide some background information about them, and especially I would like to talk about the problems with affiliating the Nuristani languages with the situating the Nuristani languages within the Indo-Aryanian group of languages. So the structure of the talk will be the following. I will first speak about the geography where the languages are spoken and also provide some background information about the speakers. Then I will talk about the Indo-Iranian branch of Indo-European and it will become apparent why this is necessary. I will say some words about how we discriminate between Indo-Aryan and Iranian within Indo-Iranian and sketch the situation in the Hindu-Kush region now. Then I will turn to the Nuristani languages themselves and introduce some interesting features of the Nuristani languages. Subsequently I will try to situate the Nuristani languages within Proto-Indo-European and Proto-Indo-Iranian and try to introduce how we are going to proceed at the University of Cologne in the years to come. This will be the last point. So geography. The Nuristani languages are spoken in just one province of Afghanistan and in just one valley in Pakistan near the border to Afghanistan. Here you see the map of Afghanistan and the Hindu-Kush region. The circle, the insert region corresponds roughly to the province of Nuristan which was previously called Kaferistan because it was inhabited by tribes who had a special tribal religion until the end of the 19th century. This religion encouraged people living in what is now Nuristan to live in a state of a permanent war against any other tribe, any tribe with a different religion to kill people, to steal cattle, to make slaves and so on. So in the years 1895 and 96 the region was conquered by the government of Kabul in a war and stopped the unpleasant activities. You see how the region is geographically formed. Nuristani languages are spoken by tribes living in river Veles cut by rivers into the mountain. We have only five of them. The language names Akati, Prasun, Weigeli and Ashpun and the languages are spoken by some 30,000 persons which was found out by a German expedition in 1935 but later estimations confirm the number of speakers for the 60s and 70s for the time of the Soviet occupation of Afghanistan. We have some 20,000 speakers of Akati, some 2,000 speakers of Prasun and some 5,000 speakers of Weigeli and 7,000 Ashpuns. The social linguistic situation is very different. Akati is a kind of lingua franca of the region and for this reason some Akati is usually learned by Prasun and Weigeli less by Ashpun. Prasun is not learned by anyone living outside the Prasun valley and so Weigeli is also not learned by speakers of Akati and so on. The Nuristani languages are surrounded by Indo-Iranian languages and now I lost my map, sorry. This is a map made by an American linguist who used to live for some 20 years in the vicinity. The colored regions are the habitats of the Nuristani languages so the blue region here and here, this is Akati and then in between is Prasun. The green region is Weigeli and this here is Ashpun. I also, the stars are representing Indo-Iranian languages spoken in the region and the half moons Iranian languages. Here is Monji and Vahi in the Badakhshan and here is Pashto which is spreading for some two generations in the region. So we see the Nuristani languages are surrounded by Indo-Iranian languages and they are generally also believed to belong to Indo-Iranian too. So the next point which has to be addressed is what are the Indo-Iranian languages? The Indo-Iranian is actually a branch of Indo-European. They are related to Italic, Celtic, Germanic, Hellenic and so on and the Indo-Iranian languages consist of two sub-branches, the Indo-Aryan branch which is represented basically by Sanskrit which is attested since late 13th century before Common Era and the Iranian languages are Western and Old Persian. The attestation starts a little bit later. Why do we actually believe that Indo-Iranian languages are related to Celtic, Latin, Greek and so on? We have actually two criteria to establish this. We have numerous matches with Greek, Latin, Celtic and so on and the lexicon and we also have inflection patterns which are very similar to those found in the Indo-European languages spoken in Europe. If you use a handbook, you would probably first find for the lexicon something like given into A, Viras in Sanskrit, Viras in Lithuanian. I just took Lithuanian but I could easily have taken a different language or a variety of languages spoken in Europe and so on. Then we have ghosts, or gods, sweet wine, bull, otte and so on. But leximes with designating such concepts as men are usually not very significant for establishing genetic relationship between languages because they are easily borrowed. Consider for instance nearly the same or partly the same leximes and the language which is usually not considered to belong to Indo-European at all. It's a Uralic language which is called Komi and spoken at the Uralic mountains as a member of the Uralic family. You see, we also have some matches between Sanskrit and Komi in leximes such as Marta, Mord, Men, Human and so on. More significant for establishing that Indo-Iranian is actually an Indo-European branch of languages are etiological matches and words with more abstract meanings such as pronouns, local and temporal alphabets and here we find plenty of evidence which is pretty unequivocal. Then we also have inflectional similarities between the Iranian and the other Indo-European languages spoken all over the old world. For instance, such as given in 3a, it's a part of the inflection of the verb to be in Sanskrit, Yankavestan, one of the most ancient Iranian languages and old Latin and Gothic. You see the inflection functions very similarly and we have clear matches in the suffixes used in the different languages but also in the personal indexes and verbs such as T in Sanskrit, T in Yankavestan and T in Latin, T in Gothic and also in the overall structure of the paradigm. We can clearly see that, for instance, in the indicative mood in the singular we have the root consisting of a vowel and an S in Sanskrit also in Yankavestan and old Latin and in Gothic. In the indicative mood we have, by contrast, the root morpheme consisting of just the consonant. The vowel is not here. In the subjunctive mood the vowel appears again and we have this in Sanskrit again in Sanskrit, Yankavestan, but also in old Latin and Gothic and so on. This means that Indo-Iranian languages clearly belong to the Indo-European language family. Indo-Iranian languages constitute a separate branch of Indo-European because of, again, some innovations in them. We have non-trivial phonological innovations and we have exclusive morphological innovations. For the phonology, the most salient innovation is probably the merger of the liquids in Proto-Indo-Iranian war. We see Lithuanian is clearly distinguishing between two liquids in Sanskrit, Old Yavestan, and Do Not. A more interesting innovation is the merger of R and O, which are distinguished by the languages of Europe, such as Greek, in Proto-Indo-Iranian plain R. But the short O turns out as long R in oval-medial syllables. It's a very characteristic innovation of Indo-Iranian. As for morphological innovations, we can consider the innovation given in five. It's a special form of middle voice, second singular imperative, based off the active voice, second singular imperative, recently univobated with a pronoun with a so-called Indo-European reflexive pronoun in the accusative singular. So we have a secondary imperative, something like carry yourself, based on yourself and carry. How do we distinguish between the two sub-branches of Indo-Iranian, between Indo-Iranian and Iranian? We also use phonological innovations, such as A, the loss of aspirated stops in Proto-Iranian, and B, the development of Proto-Indo-European S into H, were not preceded or followed by a stop. Here we can see in Sanskrit two series of voiced stops are distinguished, plain voiced stops and aspirated voiced stops. This contrast is lost in a western and old Persian in the Iranian languages. Similarly, the S, when followed by a vowel, is preserved in Sanskrit but turned into H in a western and old Persian. Now we have to make a jump for approximately 3 millennia, because we don't have old texts from the Indokush region. In the region of Indokush, we only found languages attested since late 19th century, which means roughly 3 millennia later than Sanskrit and Avestan and so on. And this makes distinguishing between between Iranian and Indo-Aryan a little bit more tricky. Here we see a range of Indo-Aryan languages, which means basically descendants of Sanskrit. And if we compare the voiced stops in Sanskrit with their reflexes, for instance in Pashai, we see that basically nothing happened. But in some of the languages, the distinction, the contrast, is secondarily abundant. This must be a very recent development, but nevertheless we cannot use this feature for distinguishing Indo-Aryan from Iranian in the Indokush region as far as the situation today is concerned. The second feature, the development of Proto-Indo-Iranian S into H in Iranian languages can be used. We see in the Aryan languages spoken, in the descendants of Sanskrit spoken in the region, nothing happened to the inherited S. As I said, we also find some clearly Iranian languages in the region and what is nice about them, is that they have experienced some more innovations which make them easily detectable. It is actually an easy task to distinguish Vahe an Iranian language from, for instance, Pashai, which is a descendant of Sanskrit in the Indokush region. Now what is the position of the Nouristani languages within Indo-Iranian or within Indo-European? Before we can talk about this, I would like just to give a little more information about what we actually know about the Nouristani languages. We don't know very much, because it is very difficult to make research into Nouristani for political circumstances. At the moment, fieldwork is out of the question. It is too dangerous there. We just have a grammar of one Cati dialect written by a Russian-language Trunberg and published in 1980. We have a collection of texts for Prasun and a grammatical sketch written by Butrus and Degano and based on fieldwork by Georg Butrus during the 50s and 70s in Afghanistan. For Waigali, we just have a grammar of just one sub-delect of the Kalashun dialect of Waigali, written by Degano, which is also based on field notes by Butrus. And for Ashkun, the last Nouristani language, we just have a grammatical sketch and some recorded texts and also a world list by Georg Morganstierner, who made research into the Nouristani languages during the 20s and 30s. So, not very much. Now, what we know for sure about the Nouristani languages is that these languages belong to Indo-European. We know this again because of the lexicon and because of the grammar. Here is just some lexemes from Sanskrit and their counterparts in West Cati, Kalashun Waigali and in Prasun. As for the grammar, it is still possible to detect some very archaic features in the Nouristani languages, for instance, the well-known Indo-European supplation in the inflection of the demonstrative pronoun. It is still found in a dialect of Waigali just approximately in the same form. It is found in Sanskrit and in Gothic, a Germanic language and in Greek. And the phrases of this system are found in Prasun, which is a different Nouristani language. So, we know for sure Nouristani languages belong to Indo-European. What is our reason to believe that Nouristani languages constitute a unit? Geographically, they look like a unit, but we actually don't know this. We assume this because of some similarities in grammar and we assume this because of peculiar reflexes of series of consonant phonemes of Proto-Indo-European stops in Nouristani. First, for the grammar, here is the inflection of just one verb to eat in West Cati and Kalashun Waigali, you see the inflection is a little bit unusual. We have to distinguish between two genders. We have different forms for masculine gender and feminine. In West Cati, the distinction is encoded by the palatalization in the feminine form, the palatalization of the suffix L. And we have just the same in Cati Waigali, where the palatalization in the feminine gender is actually not, cannot be directly observed, but we see it's the palatalization of the following vowel. So this system is actually the inflected forms of the future tense in West Cati and Kalashun Waigali actually a combination of the verbal noun, an agent noun, Ita, one who eats or eating in Cati, a copula which is preserved in West Cati but has been abandoned in Waigali. So between several languages you find very clear, several Nouristani languages share very clear innovations in the morphology and the other words, this is a typological parallel for the development of taking from Sanskrit. An agent noun combined with the copula yields future tense. There are other reasons to believe that Nouristani languages probably constitute a branch, a group of closely related languages is the development of the situation with the palatal tectals in the Nouristani languages. Proto-Indo-European possesses a series of stops which are not preserved as such in any language but have to be reconstructed because of the reflexes in the individual languages, the so-called palatal tectals, G and G with aspiration. We see in Greek that these phonemes are reflected just as plain V-Law stops but in many languages they turned secondarily into affricates and subsequently into fricatives such as in Sanskrit and Avestan. We see the palatal q in Sanskrit the reflex is sh in Avestan s. For the palatal q the reflex is in Sanskrit the affricate j and in Avestan the fricative z and so on. In detail we can see what we have to reconstruct for the immediate prehistory of Sanskrit. We have to reconstruct a series of affricates as a ch, j and j and for the Iranian languages for Proto-Iranian we have to reconstruct similar affricates, s, dz and dz. Now what is so interesting about the Nuristani languages is the following. As it seems, Nuristani cannot descend from Sanskrit and cannot descend from Iranian but they preserve the affricates roughly in the same way they are reconstructed for the Proto-Iranian. So the Proto-Indo-European palatal q is reflected as an affricate z in Khati and Waigali spelled differently but the pronunciation is the same. The Proto-Indo-European palatalized j is reflected as affricate dz in West Khati and subsequently developed into dz in Waigali and so on. We see a clear contrast to Sanskrit and also to languages descending from Sanskrit such as Shina and Pashai the so-called Dadic languages descendants of Sanskrit spoken in the Hindukush region. Strangely enough, we also find other reflexes of Proto-Indo-European palatal tectos in Nuristani such as sh which is more similar to Sanskrit and nearly the same we find for instance in Pashai reflexes of Sanskrit words such as shala, shul in Pashai but these words might easily be recent borrowings from the so-called Dadic languages which means descendants of Sanskrit. Now what does it tell us about the position of Nuristani within Indo-Iranian? We now understand that Nuristani languages cannot belong to the Indo-Iranian sub-branch of Indo-Iranian. They also cannot descend from Proto-Iranian because as is preserved in them but if Nuristani is neither Indo-Iranian nor Iranian it must constitute a separate sub-branch of Indo-Iranian. This is the conclusion which is usually drawn in the literature but we know for sure that Nuristani is Indo-European because of the evidence provided by the inflection but do we actually know that Nuristani languages participated in the Proto-Indo-Iranian innovations such as the merger of Uruguay and the merger of IAO into R? If we don't know this why not assume that we are dealing with a different branch of Indo-European? The Indo-Iranian languages are not the only languages which turned the Proto-Indo... not only languages in which the Proto-Indo-European palatal tectos turned into africans. We also have languages of Lithuanian as Baltic languages and also Russian and Slavic languages and so on. It is even possible to play with this idea and to try to compare the Nuristani languages for instance with Slavonic. We already had a look at the future tense in West Karate and Kalashunvaikali Now we have a very similar construction in some Slavonic language in all of Slavonic. The semantics is a little bit different. We have a present... we have a resultative perfect but it is made out of a result... out of a noun with the same or nearly the same suffix and so on. The difference in the semantics can be bridged if we pay attention to the lexicalized formations in Slavonic. Why don't we assume that the Nuristani languages are actually a sub-dialect of Slavonic? I don't know. I certainly don't believe that but I think the task for the future would be first to extend our knowledge of Nuristani languages and dialects by systematically collecting and describing the evidence especially for those dialects which have not been yet systematically investigated. And the second point what we are going to... what we would like to start in at the University of Cologne is to develop a Nuristani historical phonology and morphology step by step by comparing dialects of Nuristani languages with other dialects of Nuristani languages and so on. And this seems to be very promising because of first why do we need more information about Nuristani? Have a look at this at the present tense affliction of to-do in Khati in Western Khati in Eastern Khati We don't know very much about Eastern Khati What is it? Is it a difference of fix? Is it a sound change during and into term? It's very difficult to believe this in one and the same Nuristani language So we have to collect more information And if this is done then we can learn more about the immediate pro-history about the more recent sound changes which must have been operative in the Nuristani languages and to understand a little bit more about them and about the configuration of the Nuristani language group For instance we already know In Khati and Kalashan Waigali the counterpart has the counterpart in Ashkun This helps to understand a very strange situation with some leg themes In Khati and Waigali we have a retroflex affricate It is corresponding to Ashkun's term What is it? But if we consider that in Ashkun the affricate Ts preserved in Proto-Nuristani turned into S we understand the situation Proto-Nuristani Ts seem to have developed a retroflex affricate Ts in Khati, Waigali and Proto-Prasun but not in Ashkun In Ashkun we have an intrusive T between Ts and Waigali This means that we have a strong change shared by Western Khati Khati, Waigali and Proto-Prasun but not shared by Ashkun It is spectacular because until very recently everybody believed that Proto-Prasun constitutes some sub-branch of Nuristani We are asked all the other Nuristani languages have to be grouped together So we are going to step by step and collect information about the most recent prahistory of the languages to be able to make informed guesses about the more remote prahistory and some day understand more about the opposition within into Iranian and in New York Thank you