The meaning of temporal distributiveness occurs either in situations in which a habitual activity is correlated with the recurrent periods of time, or in situations in which the recurrent periods of time are accompanied by an activity. The present paper is yet another contribution to a series of papers exploring temporal constructions in Polish that express the meaning of distributiveness. It focuses on the analysis of constructions such as pracować dzień w dzień, which are exponents of the so-called cyclic and distributive time.

The paper offers an analysis which aims at explaining the role that the choice of particular verbs, their grammatical tense and their aspectual form plays in the construal of an episode in narrative prose. Two sample passages from Ursula Hegi's novel Floating in my Mothers Palm will be analysed, focusing upon the textual function of the opposition between "perfectivity" and "imperfectivity" as revealed in the use of English verbs. The theoretical framework for the discussion is provided by the cognitive model developed by Ronald W. Langacker, with the principle of metonymical reference to events (as proposed by Radden and Kovecses) supplementing the strictly grammatical discussion. In conclusion, it is claimed that it is unconventional construals that are markers of what is commonly called "literary style". The principles that underlie such construals, when analysed in linguistic terms, reveal at least some of the workings of the complicated mechanism to which this vague label is commonly meant to refer.

This paper analyses the diachronic development of English HAD BETTER (and HAD RATHER) structure(s). It is argued that the original construction out of which HAD BETTER/RATHER developed and which contained the verb BEON 'be' could be substituted by a new construction employing HAD by virtue of the fact that these two verbs in a great number of contexts are devoid of any cognitive content leading to their mutual interchangeability (§ 1 ). Section 2 is devoted to the examination of the development of the construction in question. In section 3 it is shown how the mechanisms and principles of grammaticalisation set out by Lehmann ( 1982) [2002], Hopper (1991) and Heine (2003) [2005] apply in this particular case of grammaticalisation. The bulk of language illustrations comes from the Dictionary of Old English, the Corpus of Middle English Prose and Verse and the Helsinki Corpus ofEnglish Texts. The references to the actual works follow the pattern of the first two electronic corpora.

ln our paper we analysed a corpus of runic inscriptions that belong to the first period. The runic inscriptions that we chose for our analysis are basically full sentences that contain the elements we were interested in, namely, the verb and the object. The main purpose of this analysis was to find some implications as to Proto-Germanic word order. The data obtained during our analysis suggest that the Proto-Germanic word order was VO due to the fact that there is a strong tendency to place nominal objects after the inflected verb in main clauses. However, on the basis of the data concerning the word order in compound NPs, one could rather regard Proto-Germanic as an OV language. However, if one regards the position of the nominal object with respect to the inflected verb as the basic criterion for classifying a given language either as VO or OV, and treats this level as being independent of other linguistic levels, like for example word compounds, one will arrive at the conclusion that it is necessary to classify Proto-Germanic as an VO language.

The aim of the paper is to explore two sandhi phenomena of r-zero alternation, i.e. linking and intrusive r. We address the question of lexical representation of etymologically r-less and r-full forms participating in the processes. Moreover, we discuss the influence of historical r on both vowel quality and quantity and explain some developments of etymologically r-full forms which have led both to the confined distribution of r and various vocalic reflexes in non-rhotic dialects.

The paper presents the results of an acoustic analysis of temporal phonetic parameters cueing word boundaries in Polish. Durational variability has been well documented for different languages. Word-final lengthening, word-initial lengthening, and polysyllabic shortening, all predict that segments neighbouring a word boundary will differ in their durations whatever direction such shortening or lengthening should take. In the present study, we obtained two Polish sequences brat Adama versus brata dama from 24 native speakers of Polish. The obtained results point to a complicated pattern of temporal variability caused by the boundary location in Polish.