ALBANOHELLENICA N O .

2 (2000-1 )

ARISTOTLE SPIRO
Athens

THE PARALLEL EVOLUTION OF POSTPOSITIVE

AND «PREPOSITIVE» ARTICLES IN ALBANIAN

Holger Pedersen (1867-1953)

In memoriam

ostposition of definite article is among of the main balkanisms that

takes place in Albanian, Romanian, Bulgarian and Macedonian. The
problem of the article in Albanian has caused remarkable disputes in
scholarship. There is a vast literature concerning it, as for instance the re-
searches of F. Bopp, D. Camarda, G. Meyer, H. Pedersen, N. Jokl, A.
Graur, N. Mjeda, M. La Piana, S. Riza, M. Domi, E. Cabej, I. Ajeti, Sh.
Demiraj, F. Agalliu, A. Dhrimo etc., who have contributed to the solution
of this thorny problem during the last two centuries. This problem has been
displayed as a permanent debate on questions such as «What is the origin
of the postpositive (definite) article in Albanian?» «Which article is older,
the prepositive or the postpositive?» that is «Did the prepositive article de-
rive from the postpositive one or, vice-versa, the postpositive article derive
from the prepositive one?».
Based on various argumentations, several researchers put forward a
number of opinions with regard to this issue. However, there are chiefly
two opinions on the origin of the articles:
1) The prepositive article is older than the postpositive/definite one.
The prepositive article was uttered together with the preceding noun and
was interpreted as its part. Consequently, it reappeared in its habitual posi-
tion, in the original formula noun + article + adjective.
2) The postpositive/definite article is older than the prepositive one.
Postposition preceded preposition, while the latter of these derived from the
former through repetition of the postpositive article.
The first opinion was put forth by H. Pedersen and was also em-
braced by A. Graur (dealing primarily with Romanian articles), B. Bokshi,
S. Riza etc. The weakness of this opinion was manifested through the fol-
lowing question: Is it possible for the article to exist at the beginning of the
adjective without acting any determinative function? How could it become
true for the postpositive article to acquire a determinative function, since it
did not exist at the prepositive one? Which way, then, did the determinative
function feature the category of the nouns?
The second opinion was set forth by E. Cabej (1963:78). It was fur-
ther espoused by Sh. Demiraj and was silently accepted hitherto as the «of-
ficial» view of the Albanian Post-War Linguistics. The weakness of this
opinion was traced through the following question: How is it possible for
an agglutinative element to be separated and to stand out as an independent

45
 ARISTOTLE SPIRO

element, even repeated? What was, then, the need of a mechanic repetition
of the postpositive article and the creation of another element, phonetically
similar to it?

My article aims to cast a new light on the process, which gave birth
to the Albanian articles (both prepositive and postpositive). According to
my opinion the main question is not the chronological order of the arti-
cles origin (as it has been put so far; see, e.g., Demiraj 1985:297), but the
way through which Albanian articles appeared.
The word «way» entails more or less the actual process of acquiring
them, a process with which H. Pedersen was the first to deal with this proc-
ess (his example pus i thelle «deep well» is commonly known). Follow-
ing Pedersen s path, I intend to bring up this problem again and to deal
with it on a new basis.

There is a starting point, which consists of two preconditions that

might explain the birth of the postpositive / definite and prepositive articles
in Albanian:
1. The consideration of the birth process of postpositive / definite and
prepositive articles on the basis of the grammaticalization theory.
2. The detachment of the problem of the prepositive article from ad-
jectives and its transfer to the genitive case of nouns.

The process of the origin of the Albanian articles has passed through
four stages. These stages followed a single pattern, which was the structure
«noun + adpositional element + noun». The structure «noun + adjective» is
relevant to the substantivated use of adjectives. The best environment for
their substantivization was the genitive case. In the course of time postpo-
sition became the habitual position of adjectives. The frequent appearance
of the article in an almost closed category, like adjectives, gave the impres-
sion that the article was a part of the adjective. Furthermore, its occurrence
before nouns in genitive, in spite of the loose grammatical relation, left the
same impression as if the article was a part of the genitive form of the
noun.

As regards the origin of the postpositive article in Albanian it is

commonly accepted that it derives from an anaphoric demonstrative. That
stage is characterized by the presence of the anaphoric demonstrative and
could be formally named secondary adposition (Lehmann 1988:304). I am
displaying it schematically as follows:

(a) noun + anaphoric demonstrative or

(b) noun + enclitic word

46
 THE PARALLEL EVOLUTION OF THE ARTICLES IN ALBANIAN

A noun could be the governing word of another noun. The scheme of

the syntagm is transformed at this case as follows:

(c) noun + anaphoric demonstrative + noun or

(d) noun + adpositional element + attribute

The governed noun is in genitive case. I am considering noun as a

dependant term, because this is the most natural relation. The token posi-
tion of the adjective in Albanian might be before the noun, as in many other
languages, which is proved also by the possessive structures of today which
determine nouns of consanguinity (e.g. im ate, my father, yt vella, your
brother, e shoqja, his wife etc.) Being only rudimentary today, this syntax
has once been habitual in Albanian, as well as in other languages.
Another stage consists of that grammatical situation where the ana-
phoric demonstrative was linked so close to the preceding noun, that it was
conceived as a single word. We have to imagine such a situation when the
demonstrative was altered in a functional aspect into a postpositive article,
though not incorporated to the noun. The scheme at this case is transformed
like this:

(e) noun + postpositive article + noun

This is the stage of the primary adposition, where the onetime

autonomous word expresses a grammatical meaning.
Be this as it may, in a structural level the triple-element structure
cannot but be interpreted as the structure (d) noun + adpositional element
+ attribute, which thereby turns into a dominant element. At this stage the
anaphoric demonstrative has turned into an enclitic determinative article,
being at the same time a necessary structural element of the syntagm.

The third stage consists of the agglutination of the enclitic determi-

native article to the noun. The phenomenon of the rianalyzation is consid-
erable to an important degree in this case. It is not difficult for the article to
reappear as an indispensable element of the syntactic structure under the
pressure of the general pattern (d) noun + postpositive article + attribute.
Thus, the structural «vacuum», which might have been created after the
agglutination (and consequently the fusion) of the enclitic element is
avoided, only that the reappeared word in this use lacks completely the de-
terminative function. Besides, it cannot create syntagmatic relations with
the preceding noun, because there is no need for such a relation. In that
case the same grammatical means would be used for denoting the same
function.

47
 ARISTOTLE SPIRO

The fourth stage consists of that grammatical situation when the re-
appeared adpositional element becomes a clitic with so close relations with
the sequential word that phonetically sounds as a single word.

The four forenamed stages might not be conceived as strictly sepa-

rated from one another. This process took place gradually. Its essence con-
sists in the existence of one syntactical pattern (the general scheme (d)
noun + postpositive article + attribute) and not in two syntaxes (as S. Riza
1981:123 puts it).

The study of the origin of the articles begins from the name and not
from the adjective, because the category of the definiteness characterizes
the names and not the adjectives. As long as it is true, the syntax, which
should be considered first is the pattern noun + noun. The syntax noun +
adjective is secondary and relevant to the syntax noun + noun and in a
more general theoretical sense.

The syntax noun + adjective is chronologically secondary, either.

The adjective was used in the position of the noun in genitive case, there-
fore, so to say, in a suitable substantivated grammatical environment for the
adjective, as long as the genitive indicates an attribute. In the course of time
postposition became the habitual position of adjectives.

In the case of the days of the week in Albanian, it is significant to

concretize the substantivization of the adjectives after their post-positioning
on the pattern of the nouns in genitive. Today both constructions are used:
the substantivated adjective (e.g., e hene, Monday) and non-substantivated
(e.g., dite e hene, «Moon day»), though anyway in the substantivated posi-
tion of the genitive.

Whereas the prepositive article was drawn by the sequential word, it

never became an integral part of its. The reason is that the prepositive arti-
cle has conserved its inflection by repeating formally the grammatical cate-
gories of the governing noun.

In a more traditional view, the two-fold specific relation of the pre-

positive article subsumes this word better in Syntax, than in Morphology.

The presence of the prepositive article in the adjective and other re-
spective attributes, as well as its pronunciation as a single phonetic word
gives the impression of one word. It creates an impression as if adjectives
in Albanian are inflectional. This has never been contested so far, to the
best of my knowledge. This impression was favoured by the existence of

48
 THE PARALLEL EVOLUTION OF THE ARTICLES IN ALBANIAN

the possessive pronouns, where, due to their frequent use, occur fusions,
reanalyzations and re-formations, with two-fold presences of the preposi-
tive articles (cf. te tyre, their, te saj, her etc.) The majority of the offenses
of the linguistic norm, that occur today in the possessive pronouns of Alba-
nian, have as main reason the existence and effect of the general pattern (d)
noun + adpositional element + attribute (cf. lulet e jona, our flowers,
kenget e juaja, your songs etc.) This amplifies my opinion for the existence
of this general pattern, which allowed the permanent existence of the pre-
positive article as an autonomous word.
The broader is the lexical-grammatical category where the preposi-
tive article is used, the looser its relations with the article are. Thus, this
relation is looser before nouns in genitive case and gets narrowed before
adjectives, while in possessive pronouns it degenerates into fusion.

Another important question with regards to the articles is this: «How

is it possible for the prepositive and postpositive articles, which are histori-
cally so closely related to each other, are used so close to each other and are
formally so similar, to differ in functions?» Dealing with this issue I sug-
gest that the prepositive article originated from an indirect case, like geni-
tive, and its function could not be determinational. This amplifies my belief
that the prepositive article in Albanian, born in certain syntactical environ-
ments, is still today an essential means for denoting the syntactical relations
of this language. Its impact is spread out in almost all morphological cate-
gories. A grammaticalization process has constantly accompanied its pres-
ence, where it however preserved its independence as a specific lexical-
grammatical unit.

CONCLUSIONS:

The prepositive and postpositive articles derived together from the

same anaphoric demonstrative.
The existence of the basic pattern noun + adpositional element + at-
tribute favoured the parallel appearance of the articles.
The prepositive article is a syntactical element and not a part of the
structure of the successive word. Therefore, the prepositive article is not a
part of the adjectives, and equally not a part of a noun in genitive.

49
 ARISTOTLE SPIRO

REFERENCES

Ajeti, I. (1969) Historia e gjuhes shqipe (Morfologjia historike). Prishtine.

Passim.
Bokshi, B. (1984) Prapavendosja e nyjes shquese ne gjuhet ballkanike.
Prishtine.
Cabej, E. (1963) «Rreth disa problemeve te historise se gjuhes shqipe»,
Buletini i Universitetit Shteteror te Tiranes, Seria Shkencat
Shoqerore. 3/1963:78-92.
Demiraj, Sh. (1973) Morfologjia historike e gjuhes shqipe. I. Tirane.
Passim.
Demiraj, Sh. (1975) Sistemi i lakimit ne gjuhen shqipe, Tirane. Passim.
Demiraj, Sh. (1985) Gramatike historike e gjuhes shqipe. Tirane. Passim.
Graur, A. (1929) «A propos de l article postpose». Romania. T.LV. Paris,
1929:479-481.
Graur, A. (1967) «De nouveau sur l article postpose en roumain». Revue
Roumaine de Linguistique. Bucarest, 1/1967:5c.
Hopper, P. J. & Traugott, E. C. (1993) Grammaticalization. Cambridge
University Press.
La Piana, M. (1939) Studi linguistici albanesi. I. Palermo. Passim.
Lehmann, Ch. (1985) «Grammaticalization: Synchronic Variation and
Diacronic Change». Lingua e Stile 3/1985:303-318.
Mjeda, N. (1934) Vrejtje mbi artikuj e premna prones te giuhes shqipe.
Shkodre.
Pedersen, H. (1894) «Bidrag til den albanesiscke sproghistorie». Festskrift
til Wilhelm Thomsen. Kobenhavn, 248-257.
Pedersen, H. (1900) «Die Gutturale im Albanesischen». Zeitschrift fur ver-
gleichende Sprachforschung auf dem Gebiete der indoger-
manischen Sprachen, begrundet von A. Kuhn. Berlin und
Gutersloh. 36/1900:309-315.
Riza, S. (1958) «Nyjat e shqipes». Buletini i Universitetit Shteteror te
Tiranes, Seria Shkencat Shoqerore. 4/1958:3-59.
Riza, S. (1982) «Probleme te nyjave te shqipes». Studime filologjike.
1/1982:121-138.

50
This document was created with Win2PDF available at http://www.win2pdf.com.
The unregistered version of Win2PDF is for evaluation or non-commercial use only.
This page will not be added after purchasing Win2PDF.