Advances in the Syntax of DPs - Structure, agreement, and case

262 Artur Bartnik

determiners in both languages. Throughout the paper we consistently referred to Old

English se, seo þæt as well as Polish ten, ta, to forms as determiners even though some-

times they clearly function as demonstratives. Moreover, in Polish literature ten, ta, to

are labelled only as demonstrative pronouns. However, our terminology was used to

emphasize that Old English forms functioned in some contexts as demonstratives, and

in others as definite articles. Sometimes it is impossible to tell the difference between

the two functions. Similarly, we will argue that Polish ten, ta, to can be article-like ele-

ments in some contexts. Therefore our terminological choice is justified.

Allen (2006) suggests that possessive–determiner sequences, which were also

found in post-nominal positions in Old English, could well function as demonstra-

tives and articles. Consider (38) below:

(38) Mattheus, min se leofa

Matt h e w my.sg.nom.m the.sg.nom.m de ar.sg.nom.m

‘Matthew, my beloved...’

(Blickling 229.30 Allen 2006: 161)

However, se, seo, þat in their pre-nominal position could only be interpreted as articles

(cf. Heusler 1962, who gives a similar account for Old Norse). In other words, the

bracketing [Mattheus, min] [se leofa] from (38) was possible because se could func-

tion as a demonstrative. In its pre-nominal position it could only function as a definite

article because it was combined with the adjective and only articles are allowed in

adjective phrases. This seems to be a good piece of evidence for the article nature of

the se, seo, þat paradigm, at least in some contexts. For the sake of clarity, the structure

proposed by Allen (2006: 159) is given below:

(39)

Det NP

DP

his DP N

Det AP

þam leofanþegne

This analysis can be easily extended to Polish, even though adjectives are not obliga-

tory in such configurations. We can hypothesize that ten, ta, to are definite articles

when immediately preceded by possessives and this fact does not have to depend on

the presence of the adjective. This might explain the scarcity of possessive-determiner

sequences in Polish because speakers do not feel at ease when they interpret ten as a

definite article in a language without a dedicated definite article slot. In fact, this is not