1. Univers. di Padova, 1st. di studi biz. e neogr., Quaderni 5 (Padova, 1972).

2. K. G. Karyotakis, ed. G. Savidis (Athens, 1975); hereafter cited as ‘Poetry and Prose’.

3. T. Agras, in ‘Poetry and Prose’, pp. 190–219.

4. Peri, p. 5. Cf. Louis Roussel, in Nea Estia, 1971, pp. 1589–90, who wonders whether certain puristic forms in the ‘Elegies and Satires’ indicate a particularly early date of composition for the poems in which they are found.

5. See, among others, T. Agras in ‘Poetry and Prose’, pp. 216–17; K. Paraschos, (Athens, 1962), pp. 192–206; V. Varikas, (Athens, 1978), pp. 142–145; V. Rotas in ‘Poetry and Prose’, pp. 187–9; P. Karavias in Nea Estia, 1971, p. 1524; I. M. Panagiotopoulos, (Athens, 1949), pp. 103–59; K. Stergiopoulos, Oi (Athens, 1972), p. 130; T. Malanos in ‘Poetry and Prose’, p. 242; G. Seferis, (Athens, 1974), pp. 97, 431; K. Varnalis, (Athens, 1958), pp. 229–32; P. Haris, (Athens, 1957), p. 242; G. Theotokas and G. Seferis, (Athens, 1975), p. 180.

6. In Peri’s chapter the limits of what can be called demotic are defined rather strictly. Since his chapter was one of the starting points of my research, I have followed him as far as the collecting of material is concerned. However, as will become clear later in my article, in my opinion it is not right to apply very strict standards for judging linguistic aspects of the poetry of the 1920s -that is, standards which might derive from present-day demotic or from radically applied linguistic rules.

7. K. G. Karyotakis, 2 vols., ed. G. P. Savidis (Athens, 1965–6); hereafter cited as ‘Apanda’.

8. Savidis does not mention his orthographic changes in the individual notes to each poem in his edition, but he does make some general remarks about the orthographic adaptations C(‘Apanda’ A, p. 200). No change is made that would alter pronunciation. Cases where Karyotakis himself made changes when the poems were republished in collections, or where Ch. G. Sakellariadis included alternate versions in his edition of 1938, are mentioned in Savidis’ notes, except when they concern nothing but orthography. In one instance, however, Savidis has whereas initial publication (in 1920) had In another instance, Savidis has whereas initial publication (in 1914) had Neither of these changes is mentioned in Savidis’ notes. These are only incidental observations; they mean, however, that in these and similar cases we cannot be entirely sure who is responsible for a given form.

9. In order to arrive at such a distinction, I have studied the situation of modern Greek in the 1920s and have collected material for comparison (a) from the poetry of the time, and (b) from three contemporary grammars – H. Pernot’s Grammaire du grec modeme (langue parlée), 3rd ed. (Paris, 1917), Louis Roussel’s Grammaire descriptive du Roméïque littéraire (Paris, 1922, but mainly completed in 1911), and Manolis Triandafyllidis’ (Athens, 1938), and (Athens, 1941). Written from different standpoints, these three grammars complement each other and help one to gain a complete picture of demotic before and during the 1920s, the time of Karyotakis’ poems. Pernot’s attempts to give a picture of the Greek language as spoken by the educated circles of Athens, i.e. the circles to which Karyotakis belonged, which were under heavy pressure from katharevousa. Roussel’s is based on literature. Triandafyllidis’ derives from the author’s educational activities from 1917 onward. The poets from whose work I collected material are Malakassis, Ouranis, Porfyras, Polydouri, Agras, Lapathiotis, Filyras. In addition, I refer at times to post-Karyotakian poetry or prose, not because I wished to give a picture of post-Karyotakian demotic, but because I wanted a kind of’negative check’ in order to ascertain whether a form or a word was still being used by writers who are reputed to write in ‘pure’ demotic.

10. Triandafyllidis, Historical Introduction, §76–81.

11. Op. cit.

12. C. P. Cavafy, (Athens, 1963), pp. 195–234.

13. When theoretical studies discuss greater licence in poetic language as compared to other forms of writing, they usually mean a freer use of syntax (see Enkvist, N. E., Linguistic Stylistics [Paris and The Hague, 1973], pp. 22, 101 CrossRefGoogle Scholar, and Sol Saporta, ‘The Application of Linguistics to the Study of Poetic Language’, in Sebeok, T. A., ed., Style in Language [Cambridge, Massachusetts, 1975], pp. 82–93 Google Scholar). But E. Stankiewicz includes other elements as well: ‘The utilization of heterogeneous elements pertaining to different systems or layers of language often becomes in poetry a purposeful, artistically exploited device’ (in Sebeok, p. 76). P. Vlastos says in his (p. 5) about the distinction of prose and poetry: I. M. Panagiotopoulos ( [Athens, 1943], p.9) even wonders if demotic in general is still in its poetic phase; he looks to prose for deliverance from

14. The genetive is in Peri’s opinion strange and from an undocumented nominative, The nominative however, together with its other cases, is documented abundantly, in grammars, dictionaries and literature.

15. Sometimes, the ‘radical’ form clearly fulfils the need for a rhyme. Thus is used in a situation requiring a rhyme, while occurs elsewhere.

16. Concerning Palamas, N. P. Andriotis has written: Christmas, 1943, p. 243.)

17. Cf. Pavlos Nirvanas in 21 Dec. 1919, p. 850:

18. In’Poetry and Prose’, pp. 216–17.

19. Op. cit., p. 143.

20. (Paris, 1921), pp. 109–10. This second part of the Grammar deals with katharevousa.

21. C. Th. Dimaras, (Athens, 1976), p. 449; cf. Peri, p. 15.