Introduction

Since J. Milsark (1979), the definiteness effect in English existential there-sentences refers to the restriction that not all the determiners may be built into the postverbal noun phrases (pivots). He suggested that the definite article, the universal quantifiers “all, every, each”, the quantifier “most”, possessives and demonstratives should be excluded from this position as they cannot indicate the size of the set denoted by their nouns. Thus, the ability to express quantity of some kind can be regarded as a sort of access to this particular position. This restriction has become the basis for dividing the determiners into strong and weak (excluded and admitted to the position, respectively). The following examples show this difference:

(1) a. There was a man.

b. There are some/several/many/three policemen.

(2) ?? a. There was the man.

?? b. There is my uncle.

?? c. There is every man.

?? d. There are all books.

The definiteness effect has a long and rich history (Milsark, 1977; Safir, 1982; Hannay, 1985;

Reuland and ter Meulen, 1987; Lumsden, 1988; Freeze, 1992; McNally, 1997, 1998; Hazout, 2004; Sorrenti, 2015; Bassaganyas, McNally, 2020). The comprehensive overviews can be found in (Francez, 2007; McNally, 2016) inter alia.

The definiteness effect has gained a great popularity, it has been studied on the English data (Abbott, 1993; Barwise and Cooper, 1981; Beaver, Frances and Levinson, 2005; Keenan, 2003; McNally 1998; Zucchi, 1995), on the material of other languages, for example, Catalan (Villalba, 2016; Leonetti, 2008), Icelandic (Norris, 2011), Hungarian (Peredy, 2009), Spanish (Rodriguez-Mondonedo, 2007), Korean (Chang and Mikkelsen, 2005), Sardinian (Bentley, 2004), French, Italian, Brazilian and European Portuguese, some Italo-Romance dialects, Galician, Romanian (Bentley, 2013), Danish (Mikkelsen, 2002), Russian (Paducheva, 2003), as well as in a typological perspective (Mc Nally, 2016). It has been observed by many others that this restriction has some exceptions (Erdmann, 1976; Rando and Napoli, 1978; Ziv, 1982; Woisetschlaeger, 1983; Holmback, 1984; Hannay, 1985; Lumsden, 1988; Prince 1988, 1992; Abbott, 1992, 1997, 2014; Keenan, 2003; Hartmann, 2013). The above-mentioned studies are concentrated on the definite article for the most part, restrictions on universal quantifiers “all, every” in postcopular noun phrases are given in details in (Dolmatova, 2016, 2019).

Postcopular possessive noun phrases have not received much attention yet. The paper adds to the empirical basis of the exceptional cases by introducing the data via the British National Corpus. Corpus data provide sufficient number of natural language contexts to justify the possible exceptions to the definiteness effect. Milsark’s theory is initially based on isolated sentences with no or little context. Access to the pivot position was then governed by the determiners’ ability to express quantity. Possessive determiners do not express quantity but do occur in the pivot position. We aim to overcome this contradiction and to find the constraint licensing possessive pivots. We shall try to fulfil the task by analysing the contexts, describing them, classifying their types and subtypes. Correlating semantic features of there-sentences and possessive pivots via pragmatics of the whole contexts adds up to the task.

Materials and methods

The research material is presented by the British National Corpus (BNC). We opt for the BNC because it is the “oldest” (updating the Corpus was over in 1993), the smallest (as opposed to other large corpora, e.g. Corpus of Contemporary American English or News on the Web Corpus) and it does not exploit World Englishes as News on the Web Corpus does. Thus, we need it as a prototypical corpus (if we may call it this way).

As per the definiteness effect, possessives are excluded from post-verbal noun phrases in there-sentences. The corpus data do not support this theoretical prohibition. The total number of tokens is 405. We use quantitative and qualitative methods as it is a regular procedure for the corpus-based studies (Biber, Conrad and Reppen, 2004; Leech et al., 2012; Dubovsky and Zagraevskaya, 2019; Axelrood and Brodskaya, 2020; Makoeva, Tishchenko and Getmanskaya, 2021): counting, semantic decomposition, descriptive method, interpretative analysis.  We are planning to describe and explain the range of communicative contexts for the whole construction with regard to the semantics of the existential constructions and possessives.

Results

Possessives in postverbal noun phrases in English there-sentences

BNC gives 405 there-sentences with possessive noun phrases as pivots. For the most part this construction includes two main constituents: expletive there with copula and a pivot – a postcopular noun phrase with a possessive pronoun “my, your, our, his, her, their”. Usually there is no coda in these sentences. The exceptions are rare; there are 35 tokens with the coda out of 405. And these constructions are not of great variety. There are only five types of them (not many if compared with 15 types of the coda for postcopular noun phrases with universal quantifier “all”). We shall use the following notations here: EXPLETIVE stands for “there”, COPULA – “be”, PIVOT – postcopular noun phrases, CODA – the phrase after the pivot nominal, Vpp – 3d form of the verb, Ving – ing-form of the verb, Vinf – infinitive, PP – prepositional phrase.

(3) EXPLETIVE + COPULA + PIVOT +     CODA – Clause: Vpp + PP

There was his name written in the corner.

(4) EXPLETIVE + COPULA + PIVOT +     СODA – Locative (AdvP or PP)

There's your mother there.

There’s my son in the crowd.

There was my plate on the wall beside me.

(5) EXPLETIVE + COPULA + PIVOT +     CODA – (THAT) CLAUSE

There’s my boat that I made with my two hands.

There’s my Smith’s tapes you never wanted to hear.

(6) EXPLETIVE + COPULA + PIVOT +     CODA – Ving

There’s my little clerk sniggering away.

(7) EXPLETIVE + COPULA + PIVOT +     CODA – Vinf

There’s your tractor to do.

There was my family to look after.

As stated by A. Zucchi (1995:56), A. Bende-Farkas and H. Kamp (2001: 125) and E. Keenan (2003:194), the semantic role of the coda is to provide the context for the interpretation of the pivot.

Table 1. Types of coda in there-sentences

Таблица 1. Типы коды (это часть за посткопулярной именной группой) в экзистенциальных предложениях

Codas serve as pivot modifiers forming their “modifier structure” as D. Aksel’rud puts it (2016: 65). All kinds of the coda enable to clarify the reference of the postcopular noun phrases. When there is none, it is the context itself that defines this reference. According to these corpus data, the coda is certainly not a must-have for the possessive pivots.

The corpus data provide three main contexts for existentials with possessive pivots. They are introduction reading, enumeration, or listing reading and anaphoric summing up reading.

The first block presents introduction context. There are 254 passages out of 405. In fact, it is the main role of there-sentences – to introduce a new referent. The presentation of a relatively new referent is explicitly shown in the following examples.

(8) There’s my dog. It’s called Dempsey

(9) Well, there’s its name painted on it. Just up there, look.

(10) There’s your mother there, look.

(11) We followed the noise of the fighting and came to a hill. There stood a strong wooden house, big enough for forty people, and with holes for guns on every side. All around the house was a wide open piece of land; and around that was a fence, two metres high, with no doors or openings, and too strong to pull down easily. As soon as Ben Gunn saw the English flag flying over the house, he said, 'There are your friends.'' More likely to be the pirates,' I answered.

It is difficult to expect the meaning of total novelty and indefiniteness from possessives. Their definiteness is inherent in them due to their possessive nature. We can speak rather of hearer-new or the interlocutors’-new information or about the first appearance of the person/object referred to in the postcopular pivot phrase. So, we can call it “introduction reading” when the speaker introduces a person/object one way or the other.

There are also four other readings in this group. We unite them into one type as they differ from the mere introduction reading but share similar shade of meaning. Persons/objects are introduced into the communicative situation (sometimes in flesh, sometimes in the virtual status of being my good/clever/etc someone), it may be even their first appearance or at least labelling but they are “nobody-new”. We may call it “greeting/finding/caress/encouragement” type.

(12) Aye, where's my card! I didn't get, oh there’s my card!  - finding

(13) Here’s Alison (pause) there’s a good boy, there’s my good lad- caress

(14) Hello birds. Hello birds. There’s my good boy. Oh! - caress

(15) Oh right there we are close the window. Yeah. Eh eh eh eh  Daddy'll wash his hands. Just wait there. Oh, you're brushing your teeth there's a clever boy. There’s my clever boy. - caress and encouragement

(16) We got drunk together the night before Boris left. He got the cross out and showed it to me. He kissed it.'There’s my beauty,' he said. - caress

In the next passage Mrs Hollidaye greets her fruits, she is happy to see them, she loves them, she thinks about them as if they are children who need warmth and care. Surely, these melons are not unknown to her, she grows them, she sees them every day so we are not be able to define them as “Mrs Hollidaye-new”. But she sees them for the first time that particular day and thus these melons are brought to the scene.

(17) And there was that strange smell again which Dot recognized from the darkness of the night before of things dying and rotting and growing again. Now, in daylight, she knew it was not something to fear. 'There’s my little dears,” said Mrs Hollidaye, speaking to the fruits.” Melons. And coming along nicely. Just like children. All they need is a warm corner, a nice bit of sun, careful feeding – that's splendid pig manure we put in - and plenty of interest. Then they grow and grow. Ooh, they are going to be such a treat! So long as the frost doesn't get to them first.

In the above cases we deal with the first presentation of an object/person not in the whole context but in some particular status – the status of someone’s good boy/clever boy/little dears/beauty/etc.

Table 2. Subtypes of introduction reading contexts

Таблица 2. Подтипы вводящих контекстов

In all these instances of introduction reading contexts we see the meaning of location, there is some kind of pointing. For the interlocutor-new introduction reading contexts this pointing can be formed with the words “Look!”, “over there”, so we can attach the referent in question to the particular location whether real or virtual. In nobody-new introduction reading contexts with greeting/finding/caress/encouragement elements the referent is already settled in some location and the speaker points to it and picks it up at the same time. The default meaning of existence is just a companion but not the main idea of such contexts.

The second block involves enumeration contexts, there is no scale or gradation of any kind here but the listing is present. There are 120 tokens out of 405. The typical patterns are as follows:

(18) There’s my three brothers and their spouses and one of the nephews.

(19) Just taking this along and say well there’s my business plan, that's what I intend to do,  there’s my profit, and all the rest of it, plus accounts.

(20) There’s your tractor to do, there’s my motorbike to do, and so many other…

(21) I’ve got your two polo necks, there’s your hat, your gloves, your socks, your glasses.

In the next extract the speaker sees many buildings on their way to school (as they put it “many different buildings”), decides to describe some of them and starts with their own house. The phrase “for a start” manifests the first stage of enumeration process and implies the continuation afterwards. 

(22) There are many different buildings in my usual day. I pass a great variety of them on the way to school and I myself actually live, work and play in some of them. For a start there is my house. It was built about one hundred and twenty years ago as a small but sturdy bungalow. Since that time there has been a number of extensions put on it. The first of which was probably about twenty years after it was built. 

In the following passage the speaker lists all the possibilities available for their interlocutor: “staff, other work colleagues, personnel department, advisory agencies, training organizations, specialized bodies, a person with some training”. There is no quantitative scale but there is enumeration. Though, no gradation is observed: the listed objects are mixed, not in strict order, and the speaker does not give preference to any of them.

(23) Though recruitment and selection is not a particularly difficult process to master, it is time consuming. You can seek help with it in a number of ways: there are your own staff and possibly other work colleagues; and there may be a personnel department or at least a person with some training in this area. You could also look to outside advisory agencies for assistance, such as employers’ associations, training organizations, and specialized bodies for articles and literature such as the British Institute of Management, or the Institute of Personnel Management. The latter have a free code of practice on this area. 

The above extracts give prime examples of such a context even if there is no proper listing of the objects in the form of cardinal or ordinal scale “one, two, three …” or “first, second, third …”. Possessives themselves cannot form a scale as numbers and quantifiers can. For example, the phrase “all guests were drunk” implies (due to the scalar implicature) that “some guests were drunk”. But it does not work the same for possessives: the phrase “my guests were drunk” does not mean that somebody else’s guests were drunk too. This implication is possible if the hosts belong to one group: a family, friends or any other community; or in case of extreme etiquette hospitality formulas “my house is your house”, “my car is your car”, “you guests are my guests”, “your problems are my problems”. This hospitality situation does not by all means entail reverse movement “your house is my house”, “your car is my car”, “my guests are your guests”, “my problems are your problems”. This reading is not typical but not banned though. So, possessives do not create a scale, do not support scalar implication, but they may take part in an enumeration context.

The next block of the examples presents the third reading typical for the contexts with certain possessives pivots, namely anaphoric summing up reading. There are only 31 of them.

(24) 'You can't stay here alone, Theda. You will have to hire a companion.'' A companion!' echoed Theda. 'For me?' The irony of it struck her forcibly and her gurgling laughter broke out. 'There is your answer. I shall hire Benedict to live with me and leave him the place in my will!'

Theda, the girl from this extract, could not pronounce a word because of the surreality of the idea offered by her interlocutor. “Gurgling laughter” was her only reaction. It is her behaviour, not words, that constitutes this non-verbal answer. And Theda’s interlocutor derives this answer from her reaction, thus summing up the whole situation.

(25) What's your estimate of the value of the electronic gear we have aboard? Twenty million. Maybe twenty-five. A lot, anyway.There is your answer, Doctor. That thing's gone bang once already. It can go bang once again. I am not going alongside. You are. In the launch. That's expendable. The Ariadne's not. Well, thank you very much. 

(26) At the same time she didn't see why she shouldn't benefit from it, and get herself a husband into the bargain. She decided to have a bit of fun therefore -- at their expense. "Then mark this, Master Harry: if you won't take Sam here on at fishing, you'll not have me as a wife. You either win me in fair competition, or you don't have me at all, and there’s my last word on't”.

In these two passages (25) and (26) the answers are verbal, but it is still the result of the unterlocutor’s summing up. In (26) we can present this “last word” as a complex unit dividing it into smaller parts that constitute it. We can find four of them: drawing of attention (“mark this, Master Harry”), immediate goal setting (‘take Sam here on at fishing”), adverse consequences in case of a failure of the above assumptions (“if you won't take Sam here on at fishing, you'll not have me as a wife”), detailed explanation of goal setting (“you either win me in fair competition, or you don't have me at all”). And all these elements, one by one, without missing or overlapping a stage, constitute the resulting summing up expressed by the phrase “there’s my last word on't”.

The constituents of the summing up reading are not always so explicit and detailed as in the previous passage.

(27) Now, Alison and Jack were away, gone to Amsterdam. Or rather, they would now be on their way back, since they had decided (Franca could imagine the little conversation, she thought of it as a' little' conversation) to stay away only one night, instead of the three nights originally planned. Franca was sorry about the change of plan. She had said to herself, thinking of that absence, there’s my chance! But her chance for what? She could not run away. She could not.

In (27) only one clear-cut constituent of Franca’s chance can be extruded from the immediate context – Franca’s change of plan. More close tracing needs more generous context that serves as some kind of nesting doll thus revealing the constituents of Franca’s chance.

Passage (28) gives that rare example of a possessives pivot followed by a coda. It is the coda that sums up the situation and explains the meaning of authority in this particular context.

(28) ''And you're authorizing me to leave for Brussels, to handle the problem in any way I see fit? Have I understood you?'' Broadly speaking, yes. In the interests of the Service. Draw any amount of expenses you need. There’s my authority for you to do that inside the envelope. Mission unspecified, of course. For your protection.'

In these contexts the speaker sums up the previous words or the whole situation and labels them as answer, last word, notes, chance, authority to do something, etc. There is no explicit listing (though there is implicit one by all means) in such contexts and one cannot treat them as an attempt to introduce new information.

Table 3. Types of contexts for there-sentences with possessive pivots

Таблица 3. Типы контекстов экзистенциальных предложений с притяжательными посткопулярными именными группами

The data show some ordering of the contexts possible for possessive pivots. This list is exhaustive (at least for BNC). It seems that these types of contexts reflect the stages of enumeration process itself. The nature of the listed elements does not matter, whether they are simple or complex, solid objects, abstract ideas or situations.

Discussion

In the previous section we described pragmatic contexts for possessive pivots. In this section we shall assess the data correlating the obtained information and the semantics of possessives and there-sentences.

Possessives have two semantic features: the ability to express possessive relation itself and definiteness which is inherent in them due to this possessive relation (Antonova, 2020). A possessive denotes a relationship between a possessor and a possessee, the exact nature of this relation is determined by the context. So, the possessive determiner performs two roles: it indicates the presence of a possessive relationship in a given phrase and anaphorically refers to the possessor without naming it. According to J. Taylor, proper possession is the base for prototypical possession relation. John Taylor (1989: 679-680) gives the following characteristics of this relation: “a specific animate or inanimate person who has exclusive rights to access the possessed object acts as the possessor; other persons have the right to access such an object only with the permission of the possessor; the attitude of belonging is usually long-term; the possessed object must be within the reach of the possessor, in some cases the possessed object constantly or at least regularly accompanies the possessor”. In a broad sense, possessiveness implies a rather extensive set of semantically similar relationships. In the one and the same phrase, for example, "his film", the possessive relationship between subject and object can have different interpretations, for example, a cassette with a film belonging to a person, the rights to rent a film, a film in the creation of which he participated and so on. The semantics of possession relations has been given a thorough research by numerous linguists (Williams, 1982; Lyons, 1986; Partee, Borschev, 1999; Barker, 2000; Barker, 2005; Jensen and Vikner, 2005; Storto, 2005; Willemse, 2007; Barker, 2011; Koch, 2012; Peters and Westerståhl, 2013; Kolkman and Falkum, 2020).

The type of possessive relation is determined by the semantic features of the noun denoting the object of possession. There are different classifications based on various criteria. R. Quirk (1985: 321) identifies eight main genitive meanings: possessive genitive (John's father, his passport), subjective genitive (the boy's application), objective genitive (the family's support), genitive of origin (the girl's story), descriptive genitive (a women's college), genitive of attribute (the victim's courage), genitive of measure (the ten days 'absence), partitive genitive (the baby' eyes). According to the classification based on the principle of inalienable / alienable possession (Vergnaud and Zubizarreta, 1992; Vikner and Jensen, 2002) the members of the possessive relationship are semantically interdependent. An inalienable possession, which operates mainly on the relationship of kinship, part and whole, is registered on the basis of common sense and cognitive experience. C. Barker (2011, 2000), C. Vikner and P. A. Jensen (2002: 195) differentiate ‘lexical’ and ‘pragmatic’ interpretations. ‘Lexical’ interpretations are derived from the lexical semantics of the possessive NP (e.g. John's granny, John's birthday), while ‘pragmatic’ interpretations require supporting information from the wider context. Out of the context, the phrase "my train" means the train I am traveling/meet/work/possess as an owner. Any particular meaning is provided by the broader context. J. Kolkmann and I. L. Falkum (2020: 1) claim that in fact both readings (lexical and pragmatic) receive various kinds and degrees of contextual support. As per the available corpus data we may claim that the type of possession relation (of any origin) in possessive pivots does not affect their licensing in the position discussed.

Now we shall pay our attention to the next semantic feature of possessives – definiteness. Once definiteness was the moot point in linguistics. In this paper we do not aim to give a full overview of all the theories as it does not add solutions to the problem raised, i.e. possessive pivot licensing. The most representative works on the semantics of definiteness (to name a few) are those of (Hawkins, 1978; Kadmon, 1990, Lyons, 1999. Abbott, 2004). And the main question was and to some extent still is the definitive and indisputable criterion for it. Numerous works on the problem can be roughly divided into three main groups according to what they believe the licensing condition for definiteness. There are three main ideas for this licensing – uniqueness, familiarity and salience. Salience is a more general criterion; it is based on the situational salience of the referred object. This situational salience consists of all necessary information to single out the referent. The salience criterion represents a unified combination of conditions and circumstances leading to the definiteness of the expression in question. In this case, uniqueness and familiarity can serve as constituents of salience. Thus, salience can be regarded as a common notion. Bearing in mind all their differences linguists agree on the same thing: the basis for definiteness is the unambiguity of reference, i.e. the connection between a certain description and the entity it expresses is transparent and clearly tracked. As for the possessives, their definiteness is inherent in their nature. It is the anaphoric reference to the possessor that provides salience for possessives. If we pronounce “my/his/your/her/its/our/their something” we single out, point the referent. We can identify the following semantic features for possessives: possessiveness, definiteness and deixis (as definiteness inevitably leads to deixis). We accept the idea of deixis as the function by means of which a speaker relates an entity to the current speech situation. We follow the line that all determiners are deictic.

Thus, on the one hand we have “there + be” constructions which present the semantics of existence and on the other hand there are possessives with the semantics of possessiveness, definiteness (on the basis of salience whatever it may be) and deixis. We should reveal the circumstances under which all these semantic features and their carriers (existential constructions and their possessive pivots) converge in one unit. We assume that there-sentences greet only those determiner elements in their pivots whose semantics does not contradict or duplicate that of there-sentences. Existentials introduce, first and foremost, new information, that is common knowledge. As observed and as stated earlier in (Prince, 1992; Ward, Birner, 1995; Abbot, 1997; Barker, 2000; Gaeta, 2013) definiteness is quite compatible with novelty. As E. Prince (1992: 302) put this: “There-sentences do not require indefinite NPs at all: rather, they require Hearer-new NPs”. We can continue that there-sentences do not require not only Hearer-new or Speaker-new but even Possessee-new as it is seen from the analysis in previous section. It is this Possessee-new status that accounts for semantically anaphoric uses of the pivots. So, definiteness is not a barrier for the possessive pivots therefore we claim that it is not definiteness that sets the constraints for pivot licensing.

Our analysis of the corpus data allows us to describe this interaction of existential structures and its possessive pivots in the following way. An existential construction implies existence of a certain entity, a possessive shows this entity as the possessed by one of the participants and most salient in the speech situation, the speech situation itself presents enumeration context backed up with relevant linguistic and extra linguistic information. In our corpus data all these enumeration contexts are accompanied with the pointing to the location of the entity in question. The enumeration and location are parts of deictic meaning. Thus, deictic meaning of the possessives gets support from the deictic meaning of the corresponding contexts.

The abovementioned characteristics can be summed up into four items constructing the unified meaning of there-sentences with possessive pivots: existence, enumeration, location and deixis. As we do not have other contexts suitable for possessive pivots but enumeration+location ones we may assume that the licensing condition lies in the enumeration/listing contexts of there-sentences. Possessives do not duplicate, do not contradict these enumeration/listing contexts and can be freely introduced into them, they can work at every stage of enumeration process whether it is the beginning, the listing proper or the end of it (summing up).

Conclusions

Our corpus study of the licensing conditions under which possessive postverbal noun phrases may appear in there-sentences has shown that the semantics-pragmatics interface can be very important in this respect as it deals with the issue combining the semantic features of the elements under consideration with their usage in broad discourse surroundings. The paper describes a wide range of the corresponding contexts and provide some explanations in accordance with the semantic peculiarities of the two participants: there-construction and its possessives pivots. According to our findings, we may assume that the type of possession relation in possessive pivots does not affect their licensing in the position discussed (at least there are no proofs to think the opposite). Based on a study of a corpus of natural data, we have identified three types of contexts suitable for possessive pivots. They are introduction reading, enumeration/listing reading and anaphoric summing-up reading. These three contexts correspond to the three stages of the enumeration process: the start, the listing itself and the summing up. As it is shown in the paper, anaphoric contexts give rise to the suggestion that the ability of there-sentences to introduce new information is not crucial when choosing determiners for postverbal pivots. As observed earlier, the definiteness of the possessives is neither a barrier nor a facilitator in obtaining access to the position in question. Thus, we claim that it is not definiteness that sets the constraints for possessive pivot licensing. As the above study of the corpus data show, the licensing condition lies in the enumeration/listing semantics of there-sentences. Possessives are neutral in this respect and can be freely introduced into the listing contexts. It seems reasonable to say that there are no specific licensing constraints for possessive pivots. Their access to the postcopular position of there-sentences is not limited. The semantics of possessives does not contradict or duplicate the semantics of there-sentences. As possessives are scalar-neutral elements, they can support the listing contexts being at any stage of the enumeration process whether it is at the beginning, in midstream or at the final point summing up the ideas. And it is this enumeration process that is responsible for location pointing and thus leads to the corresponding deictic meaning.