Polish has no article system. This fact does not mean that the two lexical signs that in languages such as English represent the noun significate as anaphoric, recalled by the speaker (the definite article) or as introductory, approaching a more refined grasp of the notion (the indefinite article) do not exist or are impossible in Polish.

In his study Article and Noun in English Hewson (1972 : 117) makes a claim, following from Guillaume’s theory of the article, that the anaphoric (subjective) and the introductory (objective) perspectives reflected in « the binary contrasting of the values between the singular and the universal senses of the noun’s potential significate » are « inherently present in the system of the noun in all Indo-European languages ». The objective of this paper is to find out how these two perspectives, the anaphoric perspective and the introductory perspective, underlying the fully developed article systems of languages such as English, are communicated in ‘article-less’ Polish.

The problem of expressing definiteness (which is a general term used with reference to the definite article’s anaphoric properties) in Polish has attracted the attention of polonists such as Pisarkowa (1968, 1969) and Topolinska (1984) and of some contrastivists working in English-Polish contrastive studies, notably Szwedek, who published several articles on the subject in Papers and Studies in Contrastive Linguistics (1974, 1975, 1976), and recently, Kryk (1987). Pisarkowa (1968, 1969) and Topolinska (1984) examined various functions of Polish demonstrative pronouns pointing out some that in their opinion corresponded to definite article functions in other languages. Szwedek (1974-1976) argued that word order and sentence stress patterns in Polish operate as non-lexical equivalents of the English article system. In her 1987 study Deixis in English and Polish Kryk discussed ways of expressing anaphoric reference in Polish and constructed an algorithm determining, among others, the definiteness marker function of the Polish demonstrative pronoun ten (this).

In what follows I will look at the uses of the Polish demonstrative ten trying to determine if it indeed functions as a lexical marker of definiteness. I will also survey the uses of the Polish numeral jeden (one) in order to find out if the process of its dematerialization in the direction of a possible indefiniteness marker can already be observed. Finally, I will examine Polish non-lexical equivalents of the English articles postulated by Szwedek, and I will suggest that the accusative-genitive case opposition in the grammatical object function observed in a group of Polish inanimate nouns can also signal the definite/indefinite distinction.

There are a few reasons for supposing that the Polish demonstrative ten might develop into an independent lexical marker of definiteness in a future article system. The first one is historical : in many languages possessing article systems the definite articles have etymologically developed from demonstrative pronouns. The second reason is statistical, suggested by the findings of Miodunka (1974 : 44) who observes that out of all pronouns used in spoken Polish the pronoun ten has the highest frequency of occurrence. Finally, there are uses of the pronoun ten which cannot be classified as demonstrative but have, in some analyses, been ascribed the value corresponding to that of the definite article in other languages.

Examples (1b) and (2b) below illustrate some uses of the pronoun ten which have beenpointed out as having the function corresponding to that of the English definite article :

1. a. Either the well was very deep, or she fell very slowly.
b. Albo ta studnia byla bardzo gleboka, albo moze Alicja opadala tak powoli.
(« Alice in Wonderland ». Translated by M. Slomczynski)

2. a. I’ve walked around Central Park. The park is gorgeous.
b. Spacerowalam po Central Parku. Ten park jest wspanialy.
Park ten Jest wspanialy. (Kryk 1987 : 83)

The pronoun ten in the Polish translations is employed here to render the anaphoric definiteness of the nouns in the English originals. We should observe that in ex. (lb) the presence of the pronoun ten is not indispensable for the noun studnia (well) to be interpreted as definite. The translator must have chosen to use the pronoun for the purposes of stylistic balance and emphasis. (Notice the ta-tak parallel absent from the original). In ex. (2b), however, ten is obligatory. We should also remark that the pronoun can be placed before or after the noun it defines. In her discussion of the definiteness function of ten Kryk never mentions the optionality of the pronoun in cases like (1b) or (3b) below. She quotes example (3) in her book as illustrative of a typical definiteness use of ten in Polish. Example (4), however, clearly shows that the pronoun ten is not absolutely needed to render the anaphoric meaning of the English article in a situation similar to that described in ex. (3). Cf. :

3. a. Remember the punk we saw the other day ?
b. Pamietasz tego punka, którego widzielismy któregos dnia ? (Kryk 1987 : 83)

4. a. Do you remember the picture Andrew got from me at Christmas ?
b. Czy pamietasz (ten) obraz, który Andrzej dostal ode mnie na Boze Narodzenie ?

Szwedek, who in his research (1974-1976) develops Pisarkowa’s analysis of Polish demonstrative pronouns as definiteness markers, claims (1974 : 270, 1976 : 102) that their article-like function is clear only when they are obligatory. He shows that obligatoriness or optionality of the unstressed demonstrative pronoun ten ( =this ), and of the indefinite pronoun jakis ( =some ), which are employed to mark definiteness and indefiniteness respectively, depends in many cases on such factors as word order and sentence stress. Szwedek’s series of illustrative examples is given in (5)-(11) :

5. a. Widzialem jak do pokoju wchodzil mezczyzna.
b. I saw as to the room was coming in a man.
( = I saw a man coming into the room.)

6. a. Kiedy wszedlem zobaczylem, ze mezczyzna stal przy oknie.
b. When I entered I saw that the man was standing by the window.

7. a. Kiedy wszedlem zobaczylem, ze ten mezczyzna stal przy oknie.
b. When I entered I saw that this man was standing by the window.

8. a. Kiedy wszedlem zobaczylem, ze jakis mezczyzna stal przy oknie.
b. When I entered I saw that some man was standing by the window.

9. a. Kiedy wszedlem zobaczylem, ze przy oknie stal mezczyzna.
b. When I entered I saw that by the window was standing a man.

10. a. Kiedy wszedlem zobaczylem, ze przy oknie stal jakis mezczyzna.
b. When I entered I saw that by the window was standing some man.

11. a. Kiedy wszedlem zobaczylem, ze przy oknie stal ten mezczyzna.
b. When I entered I saw that by the window was standing this man1.

Sentence (5) provides context for the remaining sentences in the series. The noun mezczyzna occupying the sentence final, focal position is introduced here as new information, a fact which in the English translation is represented by means of the introductory article a. Sentences (6)-(11) describe the situation the speaker saw after entering the room. Depending on the word order in the subordinate clauses here, and the position of the noun mezczyzna in particular, as well as on the presence or absence of the definite pronoun ten or the indefinite pronoun jakis, the noun mezczyzna is interpreted as either coreferential or non-coreferential with mezczyzna of sentence (5), which in the English translations is indicated by the definite or the indefinite article. When no pronouns are present in front of the noun mezczyzna (examples 6 and 9), the difference between the definite (i.e. coreferential with ex. 5) and the indefinite interpretation of this noun depends on the word order in the clause. Sentence (6), with mezczyzna in the clause initial position, is a natural sequel to sentence (5), and there is no doubt that the noun mezczyzna here and the noun mezczyzna of sentence (5) denote the same man. In ex. (9) the noun mezczyzna occupying the final position in the clause, can have no coreferential reading with mezczyzna of ex. (5), and sentences (5) and (9) do not form a piece of continuous text. It is quite clear that in sequences involving coreference the word order is of crucial importance when no demonstrative pronouns are present in front of the noun whose coreferentiality is to be determined. We have to be remember, however, that the noun’s position in the sentence can indicate its possible coreferentiality, and thereby its anaphoric value, only in the most natural, unmarked phonetic conditions, with the sentence stress falling on the sentence final noun. From the point of view of the anaphoric definiteness of the noun mezczyzna, ex. (7) (ten mezczyzna) has the same reading as ex. (12) (Ø mezczyzna), and ex. (10) (jakis mezczyzna) equals ex. (9) (Ø mezczyzna). This means that the pronouns ten and jakis, of examples (7) and (10) respectively, are optional. In examples (8) and (11), however, their presence is obligatory for the coreferential meaning of the noun mezczyzna changes if the pronouns are removed (see examples 6 and 9). Evidently the pronouns ten and jakis here have a greater « defining » power than that implied by the word order. The conclusion Szwedek (1975 : 170) draws on the basis of the series discussed is that « since the demonstrative pronoun ten and the indefinite pronoun jakis are obligatory in some circumstances depending on the word order and the place of sentence stress, they function in a way similar to the English articles ».

Pisarkowa (1968) points out an interesting defining usage of ten in front of proper names, which seems to me to belong to the type represented by the non-demonstrative use of ten with nouns that are already defined by a possessive pronoun and thus have their extensivity closeto that of proper names. Examples of this usage are given in (12)-(14) :

12. Bylismy najpierw ogladac ten Frankfurt. (Pisarkowa 1968)
First we went to see this/that Frankfurt.

13. Ten Janek zawsze cos przeskrobie.
This/that Johnny always gets into trouble.

14. Cóz mam zrobic z tym moim dzieckiem ?
What shall I do with this/that my child ?

This usage is typical of spoken Polish. The pronoun ten here indicates the speaker’s emotional attitude towards the object he is talking about, and it translates into English by means of the demonstrative pronouns this and that. Kryk (1987 : 94) compares this usage of ten to the « emotional deixis » value of English this and that illustrated in examples (15) and (16) quoted after Lakoff (1974 : 352) :

15. I see there’s going to be peace in the mideast. This Henry Kissinger is really something !

16. That Henry Kissinger sure knows his way around Hollywood.

According to Kryk, this usage involves a presupposition that « both the speaker and the hearer share the knowledge of the facts alluded to, and that there is something remarkable about these facts » (1987 : 94). While it is true that ten in examples (12) – (14) is anaphoric in the sense that it refers to the facts or objects both the speaker and the hearer are familiar with (our visit to Frakfurt, Johnny, my son), it is hard for me to see that there is something remarkable about the people or places described by the pronoun ten here, unless we interpret as remarkable the speaker’s emotions the pronoun signals (i.e. boredom, impatience, fishing for empathy). In his discussion of the « honorific article » in English, Hewson (1972 : 114) describes occurrences of the article the with names of prima donnas and famous women as imitation of the French or Italian usage, observing that at the popular levels of speech this French or Italian article (illustrated in the French example (17) is generally translated into English by means of a demonstrative pronoun :

17. a. Où est-elle, la Louise ?
b. Where is that Louise ? (Hewson 1972 : 114)

This usage of the French definite article with proper names seems to me the closest estimation of the defining value of Polish ten in examples (12) – (14). Topolinska (1984 : 312) must have had this type of article usage in mind when she spoke of « expressive uses » of ten in spoken Polish as corresponding to « expressive uses of articles in other languages ». An illustrative example of hers quoted in (18) is worth mentioning in the present discussion for the pronoun ten here qualifies nouns that unlike proper names and proper name equivalents, have generic reference. In (18) the speaker is talking about all girls of a certain type, not of a definite group ot girls that he evokes anaphorically :

18 a. Te dziewczyny maja tych pierscionkow na tych palcach wiecej niz rozumu w glowie. (Topolinska 1984)
b. These girls have more of these rings on these fingers than they have brains.

Developing the claim that the distinction between the anaphoric and the introductory representation of a noun’s significate exists (or is possible) in all Indo-European languages, Hewson (1972 : 118) outlines a universal pattern for the genesis of the article. According to this pattern, established on the basis of a historical study of the article systems in languages with articles, the definite article appears before the indefinite when a need arises for marking concrete anaphoric references in a language. In many languages the definite article is etymologically related to a demonstrative pronoun (as was the case of the and that in Old English, which, argues Hewson, is explained by the functional similarity between the two forms which both express anaphoric values). The appearence of the indefinite article as a separate lexical entity, morphologically derived from the numeral one, is preceded by a period in which the numeral one has the meaning of a certain and is employed to denote a more particular introductory reference. If we look at modern Polish from the perspective of this pattern and review the so-termed « article-like » usages of the demonstrative pronoun ten, we can perhaps conclude that from the point of view of the article system development Polish today is at the stage comparable to Old English. The definite article-like anaphoric functions of the demonstrative ten are detectable but they are often hard to distinguish from the clearly demonstrative functions of the pronoun. Still, ten seems to be a likely candidate for a definite article in future Polish. As to the possible development of the numeral jeden(=one) in the direction of an indefinite article, Polish seems to have entered the stage when the indefinite yet particular reference is expressed by the value of a certain clearly present in the significate of the numeral one(=jeden), as evidenced by the beginning of a popular Polish song from the area of Kraków, quoted in (19) and by the opening phrase from a Silesian miner’s joke, quoted in (20) :

19. Krakowiaczek jeden mial koników siedem ; pojechal na wojne, zostal mu sie jeden.
A certain Cracovian had seven horses. He went to war and was left with one.

20. Przychodzi jeden górnik do lekarza i powiada :
Comes a certain miner to the doctor and says :

Considering the possibility of Polish developing lexical indefiniteness markers, we can observe that the indefinite pronoun jakis (=some) described by Szwedek as having an indefinite article value (see discussion of example 10) cannot be treated as a candidate for a future lexical indefiniteness marker for it has no particularizing force, essential for an indefinite article in a binary article system.

Examining the chances for the genesis of an article system in Polish we have to remember that for articles to develop it is not enough that the language has lexical candidates with extended (dematerialized) meanings and functions. For an article system to emerge the case system in the morphology of the noun has to disappear or substantially weaken, for only then the dematerialization (in the direction of a greater notional generalization) of the noun’s potential significate is possible. (For discussion see Hewson 1972 : 66-67). In modern Polish, with its seven cases, the case system is still very strong. It is therefore reasonable to suspect that at least some of the representations reflected in the article system of languages with reduced or extinct case morphology are inherent in the case system of Polish and signalled by the observed case oppositions.

In this paper I am suggesting that some type of the definite/indefinite distinction is expressed by the accusative/genitive contrast in the grammatical object function of some Polishnouns. Examples (21)-(29) illustrate two basic genitive case usages observed in the direct objects of non-negative sentences. The VP’s of the examples in question are interesting in that they can express the verb-object relation with the help of the accusative or the genitive marked object noun, a fact which makes the distinction between the functions of these two cases easier to determine. In the sentences of the first group (examples 21-24) the accusative case on the object denotes a whole, and the genitive case indicates a part. The nouns capable of signalling the distinction in Polish by means of the two cases are uncountable mass nouns (such as cukier= sugar) and countable inanimate nouns in the plural (such as jabtka = apples). Cf. :

21. a. Kupilem cukier. b. Kupilem cukru.
I’ve bought (the)sugar-Acc. I’ve bought some sugar-Gen.

22. a. Zjadlem zupe. b. Zjadlem zupy.
I’ve eaten the soup-Acc. I’ve eaten some soup-Gen.

23. a. Kupitem jablka. b. Kupilem jablek.
I’ve bought (the)apples-Acc. I’ve bought some apples-Gen

24. a. Przynioslam jajka. b. Przynioslam jajek.
I’ve brought (the)eggs-Acc. I’ve brought some eggs-Gen.

In the English translation of examples (21) and (22) the accusative case object is rendered by means of the zero article or the definite article denoting all of the uncountable object or a defined limited quantity of it (seen as a whole). The genitive case objects are expressed as qualified by the indefinite reference pronoun some indicating an undefined amount of the object. If we agree that whole implies a definitely limited quantity whereas part has no defined limits and is therefore undefined, we can make a suggestion that the accusative/partitive genitive opposition in Polish expresses the definite /indefinite distincUon, equivalent to the 0/the versus some contrast in English.

To explain the presence of the whole/part distinction in the countable plural nouns of examples (23) and (24) a view of plurality different from the usual understanding of this concept in English has to be adopted2. A plurality of small inanimate objects should be viewed here as a divisible whole, which is described as divisible because it consists of a number of individual entities into which it can be naturally divided. In this divisible whole we can always distinguish a part which consists of a smaller plurality of these objects. Polish countable nouns in the singular do not normally exhibit the accusative/genitive pattern because they are not naturally divisible in the above described sense, and so the whole/part distinction cannot be represented in their significates.

In the sentences of the second group (examples 25-29) the value of the accusative/genitive opposition in the grammatical objects is much harder to determine. The nouns capable of making the accusative/genitive distinction denote [-animate], [+countable] single objects which cannot be interpreted as naturally divisible. Depending on the verb and the form of the verb used several differences can be implied by the accusative/ genitive contrast in the direct object. In example (25) with the verb dac(=give) and the direct object denoting an object that could be consumed by the speaker, the genitive form implies that the item is one out of many others of the kind that could be consumed. The accusative form of ex. (25b) identifies the item as a definite particular object. The noun requires a definiteness marker (supplied by ten), and the whole phrase indicates that the cigarette is wanted for something else than consumption. The speaker wants to inspect it or maybe to keep it. Cf. :

25. a. Daj (mi) papierosa! Musze zapalic.
Give (me) a cigarette-Gen. I’ve got to have a smoke.
b. Daj-no (mi) ten papieros. Za mlody jestes zeby palic.
Give (me) this cigarette-Acc. You’re too young to smoke.

In example (26a) the genitive form indicates that the hearer can have one of an unlimited quantity of objects of the kind. The accusative form (26b) makes it clear that a choice is to be made between two definite pieces of meat prepared for dinner, one being a cutlet, and the other a schnitzel. The quantity of the object is clearly indicated and the object can be seen as particular and anaphoric (=the) or as a single representative of the species (A is very close to one here). Cf. :

26. a. Co wolisz : kotleta czy sznycla ?
What do you prefer : a cutlet or a schnitzel-Gen ?

b. Co wolisz : kotlet czy sznycel ?
What do you prefer : the/a cutlet or the/a schnitzel-Acc ?

In her paper The Semantics of Case Marking Wierzbicka (1983 : 263) observes that with verbs involving consumption « the genitive suggests a pleasurable snack, and the accusative suggests a large item of seriously consumed ‘serious’ food ». It seems to me that the difference here lies not so much in the ‘seriousness’ of the food consumed as in the way the amount (limited or unlimited) of it is conceived. Example (27b) shows that the accusative case can also denote a small item of food such as a wafer :

27. a. Chcialbym andrucika.
I would like a wafer-Gen.

b. Jas dostal od mamy andrucik(a) i jablko. Najpierw zjadl andrucik a potem jablko.
Johnny got a wafer-Acc or Gen. and an apple. First he ate the wafer-Acc. and then the apple.

Wierzbicka notices also that « if the speaker is concerned about what happened to the object the accusative is definitely preferable to the genitive » (1983 : 263). Her examples are given in (28) :

28. a. Ojej ! Kot zjadl mój kotlet !
Oh no ! The cat has eaten my cutlet-Acc !

b. Ojej ! Kot zjadl mojego kotleta !
Oh no ! The cat has eaten my cutlet-Gen ! (Wierzbicka 1983)
This difference is seen more clearly when the indirect object is present in a sentence, as in example (29) :

29. a. Ojej ! Pies porwat mi but !
Oh no ! The dog has snatched of me a/the shoe-Acc !

b. Ojej ! Pies porwal mi moj but !
Oh no ! The dog has snatched of me my shoe-Acc !

c. Ojej ! Pies porwal mi buta!
Oh no ! The dog has snatched of me a/the shoe-Gen !

d. *Ojej ! Pies porwal mi mojego buta!
Oh no ! The dog has snatched of me my shoe-Gen !

We can notice that when the indirect object (mi=of me) is present, the defining posessive mój(=my) on the genitive-marked object (ex. 29d) makes the sentence much worse, if at all possible, whereas its presence makes no acceptability difference in the sentence with theaccusative- marked object (ex. 29b). This fact suggests that with the genitive-marked objects the emphasis is on the experiencer (of me) and there is no room for elaborating and defining the notion of the direct object, which, as Wierzbicka rightly observes, is not so important to the speaker. With the accusative-marked objects the defining modifiers are not only possible but welcome, even if the experiencer is mentioned in the sentence, for the speaker’s attention is all centered on the object. We are tempted to conclude that what is of greater concern to the speaker is more important to him and therefore more definite in his mind, the term definite implying a more marked presence in the speaker’s subconscience. Thus it is more easily recalled and more concretly represented in his speech.

In conclusion : Do Polish cases define or don’t they ? Yes, they do define but in a way different from the usual understanding of the definite/ indefinite article opposition in languages such as English. For divisible objects, expressed by [-countable] mass nouns and [+countable] animate plurals, the genitive case in the grammatical object situation denotes an indefinite part, corresponding to what in English is expressed by the indefinite pronoun some while the accusative case evokes a whole, which is either a defined unit (including units that consist of a number of smaller entities), rendered in English by means of the, or it evokes a continuate total expressed in English by the article zero. When the grammatical object is a countable inanimate in the singular, the accusative case denotes a single anaphorically definite object or a single definite object the speaker is concerned about. The genitive case implies that the speaker views the object as one of many possible, or as an object of little importance to him.

Notes

1. – In my reading of this example the word okno in the NP przy oknie (by the window) carries a heavy stress.
2. – In English the plural -s sign does not permit seeing a plurality of numerical objects as a continuate which contains a number of objects, interpreted here as a whole. The plural -s is a sign of discontinuum, representing a function similar to that of the definite article. (See Hewson 1972 : 105).

References

• HEWSON, J. (1972) : Article and Noun in English, The Hague : Mouton.
• KRYK, B. (1987) : On Deixis in English and Polish : The Role of Demonstrative Pronouns, Frankfurt am Main, Verlag Peter Lang.
• LAKOFF, R. (1974) : « Remarks on This and That », CLS 10, pp. 245-356.
• MIODUNKA,W. (1974) : « Funkcje zaimków w grupach nominalnych wspólczesnej polszyzny mówionej », Jeszyty Naukowe UJ 48, Kraków.
• PISARKOWA, K. (1968) : « Zaimek w polskim zdaniu. 2. Obserwacje przydawki zaimkowej », Jezyk polski 43, pp. 12-33.
• SZWEDEK, A. (1974) : « A Note on the Relation Between the Article in English and Word Order in Polish », PSiCL 2, pp. 213-225.
• SZWEDEK, A. (1975) : « Pronouns as Articles ? » PSiCL 3, pp. 265-271.
• SZWEDEK, A. (1976) : Word Order, Sentence Stress and Reference in English and Polish, Edmonton, Linguistic Research Inc.
• TOPOLINSKA, Z. (1984) : Gramatyka wspólczesnego jezyka polskiego. Skladnia, Warszawa, PWN.
• WIERZBICKA, A. (1983) : « The Semantics of Case Marking », Studies in Language 7.2, pp. 247-275.