 I am delighted to be the chair of the first departmental seminar this term, and I am delighted to introduce Yerina Nikolayevna to you, our head of department, and I'm sure that you all know Yerina, and I think that you all know how research-active Yerina is, and in fact she has just finished the book, and we are going to launch next week, and it is called typology and linguistic theory, descriptive typology and linguistic theory, something like that, and relative clauses. So anyway, so and then if you just look at the website and go to her home page, you see all the publications. So at one point I thought, you know, what's the point? It's better to give you the reference to her home page, and then you can look at, I mean, she publishes an article the day, so there you are. And she has been doing that for years and years. So this is Yerina, and today Yerina will talk to us, and she's going to be talking about the expression of time on nouns, evidence from two terminates. Yerina. Thank you very much for this introduction. I have a lot of examples here, that's why I did not prepare the slides. I only have handouts, because there would be too many to put on slides. Right. So what I'm going to talk about is the expression of tense aspect mood, or mostly tense mood on nominals. And it has been quite widely accepted in linguistics that the temporal interpretation of nominals, or noun phrases if you like, is basically established through the sources outside of the relevant NP. The time at which the noun phrase is true, either coincides with the time at which the verb of the clause is true, or it coincides with some contextually relevant time. Well, if you look at very simple examples in one, the first one, every prisoner is now in jail. The time at which the noun phrase prisoner is true coincides with the time of the main verb, it's now, basically. But if you look at the second one, every fugitive is now in jail, you can see that that's not exactly true, because when you are already in jail, when you are in jail, again, you are not a fugitive anymore. So the noun phrase every fugitive was true at some time which was before the time of now. That's where we say that the time of the noun phrase is established through some contextually relevant time. But that's basically true for many languages. However, it has been claimed that in some languages the time of the noun phrase can actually be established through the source internal to this noun phrase. So it's somehow expressed within the noun phrase itself, not in the context and not in the verb. And there has been a growing body of evidence for the morphology which indicates time on nouns. I give some references here. What it actually means for us, it means we have to decide on the question of categorization. If the nominal is bearing a marker indicating NP internal time, is it a tense marker or not? And if it is a tense marker, is it the same tense marker as the tense on the verb? One would say that they instantiate the same category as just the category of tense in language. And more generally, what do these markers tell us about the way the concept of time is expressed and grammaticalized across languages? So that's basically what I'm going to look at. I have to say that this question is highly debatable, especially the question of category. There is no agreement at the moment whether the nominal tense is the same as the verbal tense. And basically there are two different positions here. The first of you was expressed by Seminole paper by Norton and Sadler who talked about nominal tense with nominal scope as they called it. And they cited a body of evidence from about 15 languages, I think, which do have the expression of nominal tense on argument nouns. And they also cited a few properties of such languages as you can read here. In such languages the nouns show a distinction in tense and it's very consistent through the whole class. And it's not restricted to nominals which function as main predicates, but it also occurs on argument nouns. And the tense is a morphological category, it's not a syntactic logic, so it's expressed by some kind of bound morphology basically. On the other hand, other people have doubted that nominals can have true tense and said that what these markers are, they're not real, they're something else. In particular Judith Tonhauser challenged this view that nominals have tense and she said that the evidence we have from the languages which have been investigated from this point of view suggests that these markers do not behave like true tenses on verbs and therefore they do not deserve to be called tense. There's something else. And her further claim is there's currently no evidence from any language at all that languages can have true category of tense on nominals, on argument nouns. And each case is highly individual. So my paper is basically going to contribute to these debates. I'm going to investigate this temporal category in Tundra Nenets, and it's very much a work in progress I have to say. I haven't reached the final conclusion on that yet. But basically what I'm going to suggest in this particular paper is that Tundra Nenets does have some kind of tense-related morphology on nouns, but whether it's actual tense or perhaps mood depends on how we define tense, whether it looks at the predicate internal to the noun phrase or to something external to it. I'll come to that. Okay, so what I'm going to investigate, I'm going to investigate the forms in Tundra Nenets which are traditionally called predestinative forms. That's just a traditional name which is used in descriptive grammars. And you can see an example in two on the next page. So T for me, or meant for me, that's the predestinative phrase. And it basically expresses a two-place relation. X meant for Y, or Y's future X. And I will say that the X, the head of the noun phrase itself, that's the actual predestinative noun, and it bears the predestinative suffix, pred, I indicated in glosses. And the element Y, I will call it beneficiary. It can be, in this particular example, it's I. And it can be expressed by the bound agreement morphology on the predestinative noun itself, the first person singular in this case, and also possibly by a freestanding pronoun. And I have to say that there are no analysis of these forms in Nenets, but there are analysis for other languages related to Nenets, or distantly related. And they all claim that what is important here is that the beneficiary can actually correspond to the goal argument by transitive verb. This you can see in three. So if you have something like, I gave you a book, the only expression of the recipient argument is by means of this beneficiary, right? So that's what they looked at and analyzed it from various perspectives. But the point here is that at least in Nenets, the beneficiary exhibits all the distributional encoding properties of the regular possessor. And it can be shown to be actually internal to the possessive NP, to the same NP where the predestinative noun is. In four, I show the structural parallelism between the predestinative construction and the regular possessive construction in this language. For example, if you look at A and B, you can see that A is your boat and in this case the possessor is expressed by the possessive agreement on the head noun and optional freestanding pronoun. And boat meant for you is exactly the same except that there is this predestinative morpheme. And so on. There is a complete parallelism between the regular possessive construction and the predestinative construction. Moreover, I'm not going to show it here, but we do have all the syntactic evidence that the beneficiary actually does not assume the argument status. It does not show the properties of a direct object or indirect object, but behaves like an NP internal possessor. So there is no real evidence that the beneficiary is the argument of the verb. Most importantly, it does not even have to correspond to the goal argument of the transitive verb. This is shown in five. If you say something like, I gave Marsha a book for you, you can see very clearly here that the goal argument is not co-referential with the beneficiary. The beneficiary is you and the goal argument is Marsha. So I conclude then that the beneficiary is not an argument of the verb. It remains internal to the noun phrase. Basically what I'm going to suggest is that the literal translation, there's a handout somewhere there that the literal translation of example three is not I gave you a book, but something like I gave a book, I gave you your future book with the goal argument remaining unexpressed. And this is perfectly possible in the Nenens grammar. This is independently attested in the Nenens grammar. You do not have to express the goal argument at all. And in this particular case, it remains unexpressed. But then there is a certain implication that it is co-referential with the beneficiary. Okay. So I'm going to look at the syntax and semantics of these forms in more detail. But one question which I'm going to address is the question of the category. So what is this beneficiary, this predestinance of morpheme? It is not the possessive agreement because we can see that the possessive agreement is expressed independently. And apparently it's not case because the predestinance of forms, sorry, exist in three cases. And this is expressed in the paradigm which is cited on the next page. So they exist in the nominative, accusative and genitive. And you can see here that the case is expressed independently of the predestinance of morpheme. Sometimes the case accumulates with the possessive agreement, but that's normal for the Nenens grammar. So this means that it's not the case and it's not the possessive. What is it? So that's the question which I'm going to discuss in the remainder of this talk. But before I do that, let us just look quickly at the syntactic distribution of these forms. You can see that there are three case forms of predestinance of nouns. There is the nominative, the accusative and the genitive. The nominative functions as the subject or as the imperative object. Optionally, this is shown in seven. And these are the regular functions of the nominative case in the Nenens grammar. So there's nothing special here. The accusative we have already seen, that's like example three. I gave you a book. So there's the accusative predestinance if it's the object. And the genitive predestinance functions as some kind of adjunct, meaning roughly S or 4 or instead. So these are examples in eight. This dog became dog meant for us. Well, here it's not exactly adjunct. It's some kind of secondary predicate in the combination of the copula verb. Or eight B, we have bought this reindeer as food for ourselves. That's where you have the genitive predestinance. So these are the syntactic distributions of these forms. And now returning to the question of category. So we have seen that these are not the case. Predestinative morphemes are not cases. And they're not possessive. So what are they? In my previous work, I suggested that they actually express nominal future tense with nominal scope. But Nenens does not have nominal tense in non-possessive constructions. So this means that what we have here is that the tense is only expressed in possessive noun phrases in this language. This actually is not unique to Nenens because it has been claimed in the literature that there are other languages which express the distinction of tense in nominals only in possessive constructions. And some people have argued that there is an intrinsic link between possession and tense. I'm not going to go into this, but there are some theory internal reasons to believe that this should be the case. So if we analyze it this way, it means that Nenens is the language which expresses tense opposition in nouns only in possessive constructions. And it's interesting to notice that in non-possessive noun phrase, the tense usually takes scope over the reference of the noun phrase itself, the property to which it refers. But in possessive constructions, we do have two semantic predicates. On the one hand, you have the property denoted by the head noun itself. And on the other hand, you have the possessive relationship. And it has been noticed in the literature that tense in noun phrases can take scope over either of these two predicates, sort of independently of each other. In fact, Judith Tonhauser suggested that there are several potential situation times which can be localized with respect to time within the interpretation of the noun phrase. You can have the time at which the whole NP is interpreted, TNP she calls it. You can have the time at which the property denoted by the noun is true of the individual expressed by this noun, TNP she calls it. And you have the time at which the possessive relationship is in the noun phrases through TP she calls it. And then it has been noticed that languages actually fall into two types depending on how tense is interpreted in possessive phrases. In the first type of languages, I mean the languages which do have the temporal morphology of nouns. In the first type, time-oriented markers can scope over either of the two relevant predicates, either TNOM or TPOS. So there is a potential ambiguity in the interpretation of the noun phrase. To show it, have a look at example 9, it's from a language called Gorani. The claim here that this noun phrase, my former house, is ambiguous. So what this example can mean is this. On the first interpretation, the speaker is seeing the house which she used to own, which means that the past morpheme on the noun only locates the possessive time. On the second interpretation, the speaker is seeing something which she owns right now, but which is not the house, which used to be a house. So for example, the speaker is looking at the ruins of the destroyed house and saying, well, these are my ruins basically. But what they say is that they say this is my former house. So in this case, the past tense only takes scope over the nominal time, not the possessive time. So these are the languages of the first type which show this kind of ambiguity in possessive phrases. But then there are languages of the second type where there is no such an ambiguity and it seems that the tense morpheme in possessive phrases only takes scope over the possessive relationship. For instance, the claim is, and Martin can correct me if this is wrong. The claim is that in Somali this noun phrase, my former students, which I cited here, which I will not pronounce, can only mean the students who used to be mine, but are no longer mine. But it cannot mean my deceased students. So in this language, the past tense morpheme only locates the time of the possessive relationship. And apparently Tendrenenis belongs to this type. It's very similar to Somali in this respect. So only the possessive relationship is relativized with respect to time. So I showed this in 10. In example, Tend, you have yesterday I gave you a book and tomorrow I will give you a book. In both cases, the predestinative morpheme on the object book indicates that the possessive relationship between me and the book will start after the time of the event. So after the time I gave the book to you yesterday and after the time I will give you the book tomorrow. And nothing else. The predestinatives can, in fact, exhibit ambiguity because the nominal time can also be located in the future with respect to event time together with the possessive relation. But the predestinative morpheme never requires that. It's some kind of possibility. So if we look at example 11, a doctor for us arrived. It does, in fact, exhibit ambiguity but a different kind. It can be understood as the individual is already a doctor, but not our doctor. And in this case, only the time of the possessive relationship is in the future with respect to the time of the noun phrase. It can also mean that the individual is not a doctor at the event time and consequently it's not our doctor. So for example, we're talking about the student who is studying to be a doctor and we know that when he finishes his study, he will be our doctor. In this case, both the possessive relationship and the nominal time are in the future with respect to the event time. However, what it cannot mean, it cannot have this interpretation C. It cannot mean an individual with whom we are already standing in some kind of possessive relationship but he's not a doctor. So that's presumably what you have in Guarani but not in this language. So this shows that T-Pos must be in the future. Whether T-Nom is in the future or not depends on the context, on the semantics of the verb and some other things. But the encoded meaning of the predestined morpheme is that the possessive relationship is in the future with respect to the event time. Wow. Okay. All right. I think, well, I haven't even started talking about the interesting stuff. Yeah, well I'll just go very quickly through the remainder of this section. Another claim I'm making here is that the speaker believes that the predestined entity is meant to become beneficiary's possession subsequent to the event described by the sentence. But predestined entity cannot denote an entity which is not meant to become the beneficiary's possession. This means that if you say something like, I gave you a book to keep or I gave you a library book, this cannot be expressed by means of the predestined of construction because the sentence presupposes that you are not going to become the possessor of the book basically. This is again shown in 12. So if you want to say something like, I gave you a book but it's mine. So basically I just gave you a book for the temporary usage but not meaning that it will become your possession. You cannot use the predestined of construction. You have to use the regular dative indirect object. Okay. Okay. I'll probably skip the next point 13 and go to the examples 14. They illustrate another point which Don Hauser made with respect to the language she investigated, namely that the temporal morphemes on nouns in this language have the change of state mean property as she calls it. This means that they do not only indicate that the relevant property is true at a certain time t but also that it is false in other times. And this is also true for Tundra Nenets because the predestined morphemes indicates that the possessive relationship is located in the future with respect to the main event and it is false at the time at which the noun phrase is interpreted. This means that you cannot use the predestined of construction to render the meaning I brought you your book which I had borrowed, right? Because this presupposes that you, the beneficiary, has been already standing in the possessive relationship with this book. And what the predestined morphemes indicates is the possessive relationship has to start after the event of the main verb. This also means that example 14a, a passport for you arrived, can only be used about a very new passport which has been issued to you and which you receive for the first time. If, for example, you already have a passport and then you send it to a consulate for a visa and then they return it back to you, you cannot use this construction to describe the situation. And the same with 14b, he brought a sister to me. Again, this cannot be used with respect to your actual sister, somebody whom you already have as a sister. This cannot only be used metaphorically about someone whom you will adopt as a sister, but not about your actual sister. This of course implies that predestinates are very infrequent, although they're not completely excluded on relational nouns which denote the entities that stand in a permanent possessive relationship, such as certain kinship terms. Again, well, that's the same point basically. In 15a, a mother for me arrived, it does not mean your actual mother, it means somebody who will become like a mother for you. And I painted an arm for you. I cannot use it on somebody, talking about somebody's arm. What it can mean, it can only mean that I made a sort of artificial arm for you and gave it to you. I thought it will be your new possession. Right. Okay, I guess I will skip the next two points. Basically summarize the temporal meaning of the predestinative as follows. This is on top of page 7. The basic tense-related contribution of the predestinative morpheme is as follows. For an entity X denoted by the predestinative NP, the possessive relation is meant to become true of X at a time t. Possessive, which is subsequent to the NP time NP, but is false at any other time prior to this t. And the time of existence of X must include t possessive. I did not talk about it because I don't have much time, but you can believe me that that's the case. And this also corresponds to what we find in other languages. Now, with all this, the question is, can we then call predestinators nominal future? And that's my main question here. I have to say that Tonhauser says that in order for the morpheme on nouns to be called tense, it has to have exactly the same meaning related contribution as temporal morphemes on verbs. So you have to compare the properties of nominal tenses with the properties of verbal tenses. And if they're the same, then we can say that nominal tenses are instantiated in exactly the same category. And she said several properties of nominal tenses, five properties actually, or verbal tenses, which serve her as a checklist for nominal tenses as well. There are five properties. And Iman, these five properties, three properties are actually relevant for Nenets. And they show that according to these properties, this predestinative morpheme can be called nominal tense. Namely, predestinators do not exhibit any lexical restrictions. That is true. They are relevant for the whole class of nouns. They do not co-occur with other tenses within the nominal domain. And the temporal modifiers may constrain their semantic contribution. This is something I shall show in 19. And this is something which is also observed in verbal tenses. So in 19, the translation is, our next year's doctor arrived yesterday. So what is going on here? There is this NP internal modifier next year's. And this modifier actually restricts the meaning of the predestinative morpheme. The doctor will become our doctor only next year, not immediately after the event of his arrival. And that's exactly what happens in verbal tenses as well. And according to Tonhauser, this might indicate that it is a true tense. However, there are two other properties which she lists and which do not seem to apply here. In particular, canonical verbal tenses do not exhibit this change of state, meaning property, which I discussed earlier. And we saw that predestinatives do. So according to Tonhauser's argument, they will not be true tenses. And canonical verbal tenses resolve the reference time to a contextually determined time salient in discourse as we saw it before. But again, predestinatives do not do that. They only localize possessive relationship in the future with respect to the time of the noun phrase itself, but not with respect to contextually given time. This, for example, you can see in 20b. The example here is when I returned yesterday, Marsha had already made me a coat. So what we have here is the reference time, the time about which we are talking, is yesterday when I returned. I'm talking about the second clause here. The event time, the time of making this coat is the time which precedes the reference time. And the time when the NP is interpreted is the same as the event time. And the possessive relationship is actually localized with respect to this time of the NP, not with respect to the reference time yesterday. So basically, the possessive relationship is supposed to start right after the event of making the coat, not after I arrived yesterday. So it looks like according to this property, this predestinative morpheme does not behave like a verbal tense, because that's not what verbal tenses do. Moreover, and this is probably even more important, predestinances do not conform to the definition Tonhauser gives for the nominal tense to begin with, because what she says is this. Calling the markers nominal tenses suggests that they behave like verbal tenses in that they contribute to the location of the time at which a noun phrase is interpreted. So basically, the assumption here is that the verbal tense locates the time at which the predication expressed by the verb is true, and the nominal tense locates the time at which the noun phrase is true. But we saw that predestinances do not do that. They do not locate the time at which the noun phrase is true. They locate the time of the possessive relationship, which holds within this noun phrase. So that's something different. I believe that if we do talk of them as tenses, then it probably makes more sense to compare them with embedded tenses, not with the tenses of the main verb. Right? Because basically what they do, they provide the temporal interpretation of the embedded predication. Notice that predestinances are never the main predicates of the clause. They're only found on argument nouns. So if we compare them with tenses found on verbs, it makes more sense to compare them with tenses found on the verbal forms which occur in embedded contexts. For example, the tenses found on non-finite forms or as embedded tenses, rather than the tenses found on main predicates. And on this analysis, what we do have here then, we have a tense, an embedded future, which locates the time internal of the predicate internal to the noun phrase, rather than the time of the noun phrase. Right? So that would be the tense analysis. However, there is more to that and I will go through the second part very quickly because this is quite complicated. But basically the point is that there are certain properties of the predestinance of phrases which cannot easily be explained under the tense analysis, even if we talk about embedded tenses. Namely, it looks like predestinance of morphemes do contribute to the interpretation of the whole phrase, even though this contribution is not really tense related. Well, the first claim here is they actually make the noun phrase non-specific. Right? Well, there are various definitions of specificity. The one I'm using here is given at the bottom of page 8, but you probably don't have to look at it right now. What I can show you is that predestinance of noun phrase is non-specific noun phrase according to the standard specificity tests. Right? For example, it cannot refer to a reference which stands in some kind of anchored relationship with an already mentioned reference in the discourse. That's one of the definitions of specificity actually. So if you look at example 21, the context here, I bought five guns and then the continuation of this context, I will give you one of the guns. Right? If what you mean is that one of these guns is actually selected from this set of guns mentioned in the previous context, then you cannot use the predestinance of construction. What you have to say is that you have to use the regular data in direct object. Right? But the predestinance of object cannot stand in the subset relation with the set mentioned previously in the discourse. That's one of the standard tests for specificity. The second one is shown in 22. Well, basically after so-called intentional verbs, verbs like look, seek and so on, we do normally find two possible readings of indefinite NP, specific one and non-specific one. So if you say something, I'm looking for a dog. Right? This has two readings. There is a specific reading when you are looking for a specific dog and there is a reading which means that I'm looking for any dog. Right? And according to this test, predestinances are clearly non-specific because the only reading you can get here is that you are looking for any dog you can find basically. You're not looking for a specific dog. And this can be tested by means of looking at the definite on offers referring back to the center. Okay, I will not go into this either. Another thing is that the syntactic distribution of predestinances actually confirms that they are not specific. It is known that only specific indefinites can function as topics in the sentence. Right? But predestinances are totally excluded from the syntactic constructions which are triggered by topicalization in this language. For example, in this language the object agreement is triggered by topical objects. And if you take a regular object, it either triggers agreement on the verb or not, depending on whether it's topical or not. So this is shown in 23a. If you say something like I broke your boat, there is this optional object agreement on the verb and this only occurs when the object is topical. However, if you say something like I made you a boat with the predestinance of object, object agreement becomes totally impossible. Okay, that's pretty strictly ungrammatical. Predestinance of objects do not trigger object agreement. And predestinance of objects do not passivize. And again we know that in this language passivization very much depends on information structures triggered by topicalization. But it's not possible to passivize the predestinance of objects. So all these tests show that predestinances are non-specific. Right? And basically what I'm suggesting here, I'm suggesting that they have property interpretation. What they mean, they denote an individual, they denote a property type complement. And it has been argued in the literature that there are certain constructions in many languages which do induce this property like interpretation. But what exactly it is depends on the language and also on the lexical semantic of the verb and on the nature of the construction itself. One analogy I could find is in modern Greek, which I'm still hoping to investigate further. In this language we do have a very special construction type, which is called future free relative clauses with WH words. And they're shown in 25. So here you have something like I have already bought what I'm going to wear at the party. Right? And the properties of these relative clauses are mentioned here on your handout. As you can see here, they are very similar to the properties of the predestinance of construction. In both cases there is a more syntactic restriction to the future tense. These relative clauses in Greek cannot exist in other tenses only in the future. There is a resistance to passivization and topicalization. There is incompatibility with specific WH question words. And in both cases this object phrase is selected by a very closed class of verbs. I'll talk a little bit about names further in this respect. So basically what the analysis of this Greek construction suggests is that the meaning of this relative clause is property like. It does not denote a particular individual, a particular dress, but the whole class of objects. And this paper analyzes the verbs which can select this type of relative clauses as verbs which alternatively can take future-orientated objects modified by the adjunct of the type S. So basically the verbs which can license the construction in 25A also license the constructions such as I have bought a dress as my Sunday dress. And this is obviously future-oriented. But my claim is that the meaning of the men's predestinance is exactly the same as the meaning of these relative clauses in Greek. So following this analysis I suggest that they actually denote the property of indeterminate number, B, Y, X. So this means that the meaning of the first example we looked at the example 3 which is I gave you a book. It can in fact be paraphrased as I gave what is meant to or what will be your book. That's what it literally means. Okay, I'll skip this as well. Alright, there's a lot more to say here. I'll try to skip this. Yeah, well returning to the question of selecting verbs. I said that in Greek these relative clauses are selected by a very close class of verbs. And this is exactly what happens in Tudrunen's predestinance as well. Basically predestinance of subjects occur only with intransitive verbs of appearances such as the verbs enter, arrive, appear, be born. We have already seen examples. But they are generally impossible with intransitive verbs which denote the termination of existence or do not bring about a possessive relationship. For example, if you look at 27 what you can say was the predestinance of construction. You can say a doctor meant for you entered a year. This is fine. But you cannot say a doctor meant for us is still studying using the predestinance of verb. And you cannot say a doctor meant for us died. Apparently because the verb here does not bring about a possessive relation between the beneficiary and the predestinance of entry. And the same concerns predestinance of objects. Predestinance of objects are only licensed by verbs of creation such as make, write, cook, and so on. The verbs which indicate the change of location such as give, bring, buy, and so on transfer on information or discovery, find, meet. However they are absolutely impossible with verbs of destruction and manipulation which do not imply the beginning of the possessive relationship between the predestinance of entity and the beneficiary. So in 29 I showed that something like I broke a boat meant for you is impossible to say using the predestinance of construction or you love your future husband is impossible to. This is because again the verb does not presuppose that the event it describes brings about the possessive relationship between the beneficiary and the predestinance of entity. Another point here is that this actually brings the predestinances and the category they express closer to the model meaning rather than the temporal meaning. Because we know that tense is not licensed, a regular tense on verbs is not licensed by the higher verb. However the dependent moods for example the subjunctive needs to be licensed by the verb. There are many different analysis of subjunctive which actually looked at the semantics of the selecting verb and tried to make sense out of it. But the point is that only the model categories are supposed to be dependent on the semantics of the higher verb. This means that predestinances in this sense are much closer to the model meaning than the temporal meaning. I'm actually suggesting here that they can be analyzed as expressing some kind of model category probably subjunctive or irrealist. And in this sense they can be compared with the verbal categories in languages which do so-called tenseless languages which do not express the category of tense as such. But divide the verbs into two classes basically realist and irrealist. So in a similar manner predestinances divide the class of possessed nouns into two subtypes realist or non-predestinance of the irrealist or predestinances. To conclude this part of the talk I'll go to the last page right now although I did skip quite a lot of argumentation here. But basically what I said in the second part of the talk is that well predestinances do affect the temporal interpretation of the possessive relationship within the noun phrase. That's true, that's what we thought to begin with. But they also affect the interpretation of the whole noun phrase. Not only of the possessive relationship but of the whole noun phrase. Even though they do not change the time at which the whole noun phrase is true. Instead they change its referential status. They make it non-specific, right? They contribute this non-specific property interpretation. And they also indicate that the epistemic agent the beneficiary is not committed to the pragmatic existence of the entity. I did not talk about that but believe me that's true. So yes this means as I said that the meaning of the predestinance of construction contributes to the interpretation of the possessive NP. But it's comparable to the meaning of the dependent mood, subjunctive or irrealist which normally indicates the lack of commitment to the truth of the proposition on the part of this some kind of epistemic agent. Which is normally the subject of the main clause. So from this point of view it can be called mood, right? So going back to the question of category now. Well what do we want to say then? Is it tense or is it mood? So we did see that it does have some properties of tense but it's probably not the regular tense. But we also saw that it has some properties of mood. So I believe that the answer to this question depends basically on how we define the nominal tense. Whether the nominal tense is supposed to look inside the noun phrase and take scope over the predicate which is internal to the noun phrase. Or the nominal tense is supposed to temporarily locate the whole noun phrase, sort of look outside. And as I said earlier there are two different answers to this question on Tonhauser's account. The nominal tense must affect the whole phrase. So on her account predestinances are not going to be nominal tenses. So indeed this makes her general claim true that we do not have evidence from any language that there are tenses on noun. If you define them that way. But even on Tonhauser's definition predestinances can be called nominal thumb. Except that they're not tenses but moods because they do affect the whole interpretation of the noun phrase. And as I said they can be compared to the thumb system in the languages which do not have tenses such but divide the class of verbs into realist and irrealist. So that would be one possible solution. On the other hand if we do not follow Tonhauser's definition of nominal tense but rather go with Norlinger and Sadler who say that the tense on nouns does not actually have to affect the whole noun phrase. What is required here is required that the tense affects the temporal interpretation of relevant to the noun phrase in some way. And on this definition it's perfectly possible to apply the category of tense to the possessive relationship only. So on their definition I believe the predestinances can be classified as nominal tenses. But I also have to notice that a similar understanding is found in other literature. For example in papers by Le Carme who analyzed Somali. And she actually argues for exactly this second understanding of nominal tense and moreover she says that there is no principle difference between nominal and verbal tense. Basically tense is just the category of human language and it applies to the phrases of different types. Well she follows some syntactic assumptions which I do not necessarily agree with but anyway that's something to look at. So she says that nouns basically contain nominal so rather noun phrases contain the syntactic note, syntactic position called tense position t. And then of course it's realized on a language particular basis. In some languages like Nenets and Somali there will be the actual morphological marker merged with this position in other languages this position would be phonologically empty. But she also makes another interesting point namely that the tense, well she argues for some kind of parallelism in the structure of the clause and the noun phrase. And says that if in clauses you do have some kind of chain, well the syntactic dependency between the tense phrase and the whole clause, the complementizer phrase. In the same way in noun phrases you have the relationship between nominal tense and the D note. Sort of the higher note in the interpretation of the noun phrase which is responsible for the referentiality. So basically what she says is that there is an intrinsic link between the nominal tense and the referentiality of the clause. And she cites very arguments from Somali. In particular she looked at the modal effects associated with tense morphology or nouns in these languages. And notice that in that language in Somali the modal distinction which is associated with nominal tense is the one related to evidentiality. Basically it translates into the position visible versus non-visible which kind of makes sense for nouns because that's how we perceive objects as visible as non-visible. So in the languages where nominal tense has some kind of evidentiality meaning the past tense is usually associated with non-visibility on nouns and the present tense on nouns is usually associated with visibility. What Tandrin Nenets shows though that there is another modal dimension to the temporal morphology on nouns. It's not necessarily evidential in nature. It's not about visibility or non-visibility. It's rather about realist versus e-realist interpretation. But it also shows the deep connection between tense on nouns and the temporal and the referential interpretation of the whole noun phrase and the sort of modal contribution which this morphology makes. So that's my basic conclusion at that point. But I guess the basic point is the one which Tonhauser makes namely that well there are this time oriented markers or nouns in various languages and many people say that well they do instantiate the category of tense. But if you start looking at them more closely it looks like they have very, very different meanings in different languages and to which extent they do instantiate one and the same category is of course a big question. But what we really need we really need to look very closely at the semantics of these markers across languages before we actually can make generalizations on the validity of nominal tense as a grammatical category. Yeah, I guess that's all I wanted to say. There are references, well a lot of them they are available up on request. I didn't put them here on the handout. Thank you. Thank you. Half an hour for questions. Thank you very much. You start by saying these are thumb categories and you look at the tense and you say tense is quite a right thing and then you look at the move. But I was wondering whether you look at us that's all. Yeah. Some of the discussion I had to me sounded that, you know, a special distinction that was curled about. Yeah. The distinction between locating an event in the cycle of time as tense but looking at the internal constituency of the event in terms of aspects. Yeah. Sensitivity, lexical, semantic distinctions. We have that with aspect in particular when you look at axioms and verbal aspects. And what you don't get here I think with the modality with the realist, the realist is the change element. And I think again their aspect is probably more interesting. There is work by a former PhD student Peter Nichols who looked at tense and aspect and you see Swati. And there is, it looks like aspectual categories which look similar to that. There is a specific approach to change because of alterative terminology. But it is, you know, it's events which have said that it didn't hold to the past. Yeah. That's exactly what Tom Hauser says about Guarani. She actually analyzes this temporal morpheme in Guarani as an aspect marker. At least in some of her papers and exactly based on this change of state property because true tenses do not have this property. Only aspects do. And I think this will work here. But again there is the question of what are we talking about? What relationship are we talking about? Are we talking only about the possessive relationship or are we talking about the interpretation of the whole noun phrase? If you talk only about the possessive relationship, I believe you can probably argue that it's something aspect like. But if you look at the whole phrase, then I cannot really see aspectual characteristics of the meaning of the whole phrase. It's much more comparable to the mood, as I was trying to say. But yes, you're absolutely right. I would not contradict the last point I made here following Le Carme that tense could be a universal category typical of nouns in the same way as verbs. Because we know that the temporal interpretation of clauses or verbs comes from several sources. It can come from temporal morphology on verbs. It can come from lexical temporal modifiers. It can come from the aspectual information. But it all merges in her analysis at least. It all merges within this temporal note in the structure of the clause. And presumably she could argue that the same thing happens in noun phrases as well. So it all depends on your definition. That's I guess the basic point. But you're absolutely right. It does have some aspectual characteristics. You're presuming there is a distinction in some kind of universal category between noun and adjective? I guess. Right. Okay. Oh, that was the question. Sorry. Yeah, I guess. This is an endangered language, isn't it? So is it under the influence of Russian? I presume that's been quite a bit of an influence of Russian. How widespread are these predestinative markers? How often are they now replaced by a sort of more relative clauses or other type of instructions of Russian type? That's a good question, actually. I guess what I'm describing here is the language of the older generation and more competent, so to say, speakers of the language. That's true that younger speakers and speakers who are less competent would use the alternative construction, which is perfectly available in this language. It would simply use the native indirect object, for example, with die transits of verbs. I mean, I do not have any statistics, but I do believe that we can notice some kind of decrees in the usage of these constructions. Very interesting talk. I'd be interested to hear something more about this supposed intrinsic link between possessive collection and tense. Oh, well, that's very theory-internal, as I said, and I'm not really competent to answer this question with any depth. But basically, the idea here is that, as I said, the principle is that there is a certain chain, a relation between the D-note and the tense-note, and apparently, and the claim here is that somehow tense is only found on referential and specific DPs, which do have the D-note and possessive note filled in. On the other hand, the non-specific DPs, the claim is, cannot have independent temporal interpretation. Okay. Only the specific DPs do have the tense, which is independent on the tense of the main verb. So that's how it works. Also, there's no kind of a tense, so it's purely formal. That's a purely formal account, but the claim is, I mean, in neutral terms, the claim is that only the specific DPs can have independent temporal interpretation. Then it actually shows that there, that might well be true what they said, but then it actually shows that the reverse is not necessarily true, that as we can see here, non-specific and peace can also have an independent temporal interpretation. Have you read what John Said has written about these sentences? No. I guess it's you, because he just taken what the count says about this whole notion of tense, which I do sort of agree with him, which is there, because there's more sort of going on in the healer, more from a sort of pragmatic perspective. Right. And the way that these are used, rather than a sort of syntactic or a sort of sentence internal, sort of tense way of looking at it, is you do, you can have both this sort of remote and non-remote sort of determining what's in the sentence, and really it's sort of dependent on the sort of pragmatic... Yeah. Yeah. Okay, thank you. And you can guess it even with, I mean, it's... The classic example is like literally saying the time, what time is it? If you're in Somali to say, you know, it's five o'clock, you use the word for five and the determiner. So five o'clock, if I'm going to meet you at five o'clock later today, I say shantap. If I make you at five o'clock and it's already seven o'clock, you're talking about five o'clock in the past, I say shantin. If you're saying about now, you say, what time is it? And I say it's five o'clock. I say, why shantin? I'm using the past. Her past want to actually sort of say what time it is now. And it's... There are all sorts of things going on, I mean, in terms of semantics with adjectives, for example, there is... You do have the past sort of tense of the verb to be with adjectives with the thing with this sort of remote e-ending. But there is... There's sort of... There's quite a lot more to it. Yeah, yeah, yeah, yeah. It has to be done on the Somali. Sure. And it's very useful and... Sure. But I'll pass you on, John, saved the article. Thank you. It gives you a bit more sort of... Yeah. The usage of these things. Thank you. I'd like to come in again with the shantar in shantin. Isn't this also used for the icsis, for example, this and that? Well, it depends on... I mean, in terms of English, I mean, when I'm teaching, sometimes I say, because this e-ending, this remote, is sometimes... For example, if we were talking about, you know, the paper that Irenia gave today, you might say, if I'm talking about paper... Oh, that paper Irenia gave, that paper. That's where Somali uses this e-ending. But it's... For this, when you're sort of... Sort of running to the, you know, this paper, this bit of paper, then you would use a different one, which is the sort of determinate shantan, wakad dana. And you've got a specific sort of this, but the e... I mean, it's... Sometimes it's the equivalent of what we want to say, that or even this. Yeah. But it's not a particular sort of... But isn't it this visible as opposed to invisible, as I mentioned at the end? I mean... Or partly at least. Now, you could... I mean, you can use either of the determinants that add in the e for visible and invisible. At the time, there is this sort of thing that we were talking about earlier, that sort of passive sense. Yeah. It is invisible. Which, yes, yeah. But it could be... I mean, yes, I mean, in certain cases it would be... Yeah. It is... I mean, tend to the samaritans, there's a lot more to be done when they change this. And also, how they relate, when you have nanfaces in that position, they sort of... More than one nanphrase, basically, you know, one man with these things in a phrase, where they sort of interact between... Yeah. Or you can have both, and what... You have both, yeah. Right. Yeah. I mean, it's... Also, it's... I mean, because there is a thing about it not being on the... I think she... I can't remember... Not being on the sort of spatial data, this and that. But there is an ending which... No, it's actually sort of... It does seem to be there. She does have it on, you know, this one that we were talking about. It's rather than just this. Thank you. Yeah, a bit of fascinating stuff. Yeah. I just have a quick question about... Tom has his... One of them is on page seven. They can't... They can't co-cur with private answers. Yeah. Yeah. It's just... It's not about... I was just thinking about English. You could say something like, my ex-wife to be... Yeah. But that's why it's not a tense. So that's why it's not... Well, she claims that verbal tenses... I mean, these are the purposes of verbal tenses which she's trying to apply to the noun phrase, right? And she says that two verbal tenses are true verbal tenses. Do not co-cur. I'm not actually sure that it's 100% true for all languages, but that's what she's saying, right? So you can't be past and future at the same time. Although, well, I mean, historically... Historically, you can, obviously. Yeah. Yeah. I don't know. That's what she says. No, but in English, in languages like English, you have these morphemes like ex and to be or modifiers like former or future, my future husband or whatever. And the claim is that they do contribute to the temporal interpretation of the noun phrase, but they're not grammaticalized tenses. And that's why they're much more flexible and much more lexicalized. Right? Again, this would be compared... In this view which I expressed on the last page, they would be... The claim would be that English, like all languages universally, presumably, does have the category of tense in the noun phrase, except that it's phonological and null. This tense note is phonological and null, but what you can do, you can express the temporal relationship within the noun phrase by means of these lexicalized morphemes. That would be comparable to a language without grammatical tense on verbs, but where you actually express temporal contribution of the sentence by means of adverbs, let's say, yesterday, tomorrow, or something like that. Yeah. That's what these people would say, I presume. I'm not saying that I agree with that, but that would be the claim. Sorry, Nancy. I'll come back to you. You know, I'll briefly look into your hand, but you have grammatical examples where you can't use a form with the predestinate marker. But you don't have to have examples where you have to use the predestinate marker. What are the contexts where you may have to use it but what happens if you don't use it? Is it the case that you cannot use non-American nouns in context where you also could use American nouns? Well, I don't think there is a context where you must use it, but in the same way that there is not a context where, well, more or less, where you must use, let's say, past tense in English. You use it when you want to denote the past. And that's the same here. You use this construction when you want to place the possessive relationship in the future and when the whole phrase is non-specific, basically. So it would be called, I mean, if it is an inflectional category, and I presume it is an inflectional category, that would be called inherent inflection rather than contextual inflection. It does not depend on the context. It depends on the sort of internal semantics of what you actually want to express. Why would you use a future in English? Well, because you want to locate the event in the future. But I think that that's part of what I'm asking. So it's not the case that unmarked forms would express the same interpretation? No, I don't think so. So in that case, that would make the interpretation constant. Yeah. Yeah. Yes. Well, I don't know. No, I was just thinking, in general, coming back to the general notion of tense and modality, there are, you know, in Russian, there are sentences that have no tense or model indicators whatsoever. They have one for nine verbs, it's the main verb of the sentences. So maybe we're going to look into those as well, in general, to look at tense and modality as such. Where does tense and modality appear, which it clearly does appear in no sentences. That would be interesting to maybe just to broaden the whole notion of tense and modality because exactly what you said seems quite restricted from what Tom Hauser kind of tries to box it up. Yeah, yeah. So this is this, but how can you, when we have, we generally have in major, in big languages, you know, spoken with people. I mean, but that's something, we don't have to go far. We don't have to go into... Yeah. But I guess the question here is the difference between the grammaticalized tense and the temporal interpretation. Obviously, in every language, every clause is somehow temporarily interpreted. Well, more or less every clause. But not every language has grammaticalized category of tense. And what she's looking at, she's looking at the grammaticalized categories. No question, Irina. I didn't get why 29 is not good. Yeah, sorry. I had to go very quickly through this. Yes, I know. I thought that I was exactly... Oh, it's because, well, I actually skipped this part on the previous page. If you look at the page 11, what previous page here, what I'm suggesting here, there are heavy restrictions in the middle of the page. There are heavy restrictions on the lexical semantics of the verbs which select predestinance of subjects and objects. Well, the first one is what I talked about. The verb co-occurring with the predestinance of argument must denote an event which brings about the possessive relationship. Okay? And if you say something like you love your future husband, you know, loving someone does not presuppose that you are going to start the possessive relationship with this entity, I guess. I guess that's very true. But that depends on the lexical semantics of the verb, which kind of denote that you are already standing in the possessive relationship sort of some kind of pragmatic association with the entity cannot be used with the predestinance of object. Okay, thank you. So you found your future husband is fine. Yeah. But you love your future husband is not fine. Okay. Because basically when you say you found, you describe the first encounter with the object, right? But not when you love. We can thank you in that. Thank you.