Rebetiko > Arghile mou > Hasiklidika Introduction

Hasiklidika Rembetika

By Suzanne Aulin & Peter Vejleskov

1 Introduction

Gentlemen, our objective is to present a comprehensive, reliable and argued anthology of pre-war rebetiko songs that have as their main subject hashish smoking and use of hard narcotics. Then we considered it advisable to include a detailed analysis of the styles and issues of the songs.

We recognize we owe a debt to the book by Stathis Gauntlett, "Rebetika - Carmina Graeciae Recentioris", which is a revised publication of a doctoral thesis that was submitted in 1978 to the University of Oxford. As far as we know the present study constitutes the first presentation in Greek of the basic contribution of Gauntlett to the study of rebetiko song.

In the search for songs we realised that the texts of the anthologies that were published in Greece are generally questionable. Consequently we preferred to include fundamental songs which we transferred ourselves from republished original disks. (1) The work of Gauntlett also contains an extensive anthology of songs. From the songs that we heard and those Gauntlett has transferred from old disks and from tape recordings, we realised that the quality of his transcriptions was so good that we decided to include his nine songs that we did not have the occasion to hear ourselves. (2) Gauntlett's book is also distinguished in that it reports the instrumentation and, whenever it is possible, also the singer. In this way it gives more complete documentation than the other anthologies.

Following the example of Gauntlett we separated our songs into two categories with base the musical element, and we dared use the name "school" for the two teams of rhymester-composers that represent these two categories. In the school of Café-Aman musical instruments such as the violin, the santouri, the kanonaki and the accordion are used, and songs are sung by high men's and feminine voices. On the contrary, in the Piraeotiko school, songs are sung by nasal, throaty men's voices and the song is accompanied always by the bouzouki or baglama, or even the two instruments together.

Globally our anthology includes 43 different Café-Aman songs and 28 Piraeotiko songs. Songs A 21, A 24 and H of 1-5 are unknown composer, and the final five are combinations of traditional verses. A 13,.28 and 21 exist in two variants and A 33 in three variants.

Altogether, our anthology includes 43 different songs of Café-Aman and 28 Piraeotiko songs. Songs A 21, A 24 and H of 1-5 are by unknown composers, and the final five are combinations of traditional verses. A 13, 28 and 21 exist in two variants and A 33 in three variants.

That we include exclusively sound recording implementations of songs that have as subject hashish usage and use of narcotics, facilitates, up to a point, the chronology. For the Piraeotika school, we know the disks were recorded in between 1933 and the end of 1936. For the Café-Aman school, we have the same terminus ante quem; we do not have however one sure terminus post quem, because the older list of disks that we had our disposal is from 1931 and when the hasiklidika rebetika phenomena occurred they had been recorded on disk in Greece already from 1925. (3)

In the anthologies there are enough songs with main subject hashish, that claim that they have come out on disk. Certain of them we found in the lists, but no in available republication, and these we preferred to omit. We report the titles of these songs and our source in annex A. Moreover we omitted certain songs that simply mention the subject of hashish usage, whether heard on disk, or not. The titles and sources are reported in annex B.

We did not include in the anthology songs recorded in the USA for the domestic market, because they were created in another society and cannot teach us anything about the environment of hashish using in Greece at that time. For the same reason we omitted songs of A. Kostis.

2 Definition of rempetiko songs

Gauntlett examines in chapter 1 (p. 3-42) a lot of different perceptions for what rebetiko songs are. None of the work that it reports offers an explicit definition. For this Gauntlett declares (p. 43) that his main objective is via an analysis of songs to reach a reliable definition. He includes in all his analysis "rebetika" apart from the entechno popular songs and amanedes. In chapter 5 (p. 190-94) he concludes his analysis with base that genuine Rebetiko is a type of urban song that is characterized by its depiction of "marginal life"(4). The definition of urban song, however, presupposes the two traits of anonymous composition and oral delivery. Consequently, for Gauntlett the songs of our own anthology cannot be named rebetika with the narrow meaning of the word. As to the big majority of the remaining "rebetika", he characterises them as commercial derivatives of Rebetiko and says that they should more rightly be called "rebetika-like songs". He himself adds however (p. 193) that his definition is exposed to criticism for meticulousness. In this we agree. However we admit that his definition is at least reliable. Thus and differently, and in the "rebetika-like song", as in what is called Rebetiko, the two prevailing subjects are, love and the "marginal life".

3 About Rebetikology

3.1 Introduction

In the history of research above in the rebetiko song the two more important contributions are the anthology of Petropoulos 1968 (PJ) and the thesis of Gauntlett. Petropoulos was the first to make accessible to potential students a comprehensive collection of rebetiko songs. The publication of his anthology did not contribute however to a direct blossoming of scientific study of rebetiko. Seventeen entire years would pass before Gauntlett published in 1985 the first serious, completed and sufficiently argued treatise. He was the first to inform us of the origin of all, without exception, of the songs that he reports, and simultaneously expresses the required reserve for the doubtful sources. Also, Gauntlett is the first university tutor that discusses in detail the elements on which he bases the chronology of the songs. We will return to the two subjects of chronology and reliability of anthologies below.

3.2 The opposition to the hasiklidika songs

In the present study, as we said, we limited ourselves to dealing exclusively with songs that mention hashish usage or use of hard narcotics. The fact that so many rebetika songs have as their subject hashish usage, the underworld, prison, etc made the dictator Metaxas impose on them a strict and effective governmental censorship (Cf. below, heading 4.2). Later, the insistence of rebetiko on the description of marginal life had as result for many years made the subject unworthy of serious scientific research, at least with regard to the content of texts.

The third chapter of Gauntlett's book (p. 3-42) constitutes an in detail presentation of previous discussion of rebetiko. And his general appraisals and his individual judgements find us more or less in agreement, and we do not have to add a lot of things.

The discussion began in 1947 from the columns of "Rizospasti", at a atime, that is to say, where the tradition of rebetika was still alive.(5)

Certain critics claimed that rebetiko was a foreign element to the Greek culture, that its music was "Turkish" and that the contents of its verses were immoral and dangerous, likewise, for the youth. Some went so far as to demand the complete prohibition of the distribution of all kinds rebetiko song. A single extract will be enough in order to demonstrate the character of this discussion: "...the sickening atmosphere of rebetiko's hashish-using pornographic songs" (6). That "Xenos" calls the rebetiko "pornographic" shows that he obviously does not know what he is talking about.

Other observers reacted stressing the virtues of songs, and likewise the music (7). It deserves however to be remarked that many supported Rebetiko, and then later suppressed the fact that many Rebetiko songs reported hashish usage and other illegal activities (8). One of the most recent examples is the "Rebetiko Anthology" of T. Skorelis. In his introduction (Volume A, pages 11-18) hasiklidika songs are only mentioned once (p. 16), and there his unique aim is to underline that they did not write hasiklidika songs on the records in his possession. In his anthology however, countless hasiklidika songs are included, and consequently there are no clues on their deliberate exclusion from this.

Previously Skorelis and Ekonomidis, in order to "justify" the fact that G. Rovertakis and K. Roukounas wrote and sang hasiklidika songs, had felt the need to speak for hashish usage in the introductions to the biographies of the two artists (GR and KR).

And with regard to the presentation of hashish related songs the anthology of Petropoulos is distinguished, because he himself does not try to hide or "justify" anything. On the contrary, judging from what he says in the introduction to the anthology and in the remainder of his work, one could almost suspect that Petropoulos wanted to give it excessive importance in the relation of Rebetiko with the "group" hash songs and the rest of the "underworld" (9). But the particular choice of songs does not cause the impression that Petropoulos left out certain subjects because of the charm of others. On the contrary, it appears that in his anthology he includes any verses he finds that would even have a surface resemblance to Rebetiko song. This is in effect the same for his big second publication (P2). In this are found songs that even Petropoulos would admit cannot be characterized as Rebetiko songs (10). Moreover, the majority of the songs in the anthology are love songs.

3.3 Critical examination of attribution of songs in the anthologies.

Petropoulos separates the songs into various categories of main subject: "erotica", "the underworld", "hashish songs", "prison", etc. Such a distribution of material is genuinely useful. In a lot of cases however we disagree with the particular classification of songs. For example, the criteria based on which it places a song in the category "hashish related" appear many times completely subjective, and they are based on the estimate of the atmosphere of the song. Thus one finds in the category "Hashish songs", songs that do not make nor touch upon the hashish, e.g. P2 page 131 b 1 and g 1. Conversely, the song "Conversation with Charos" that deals almost exclusively with hashish is placed in another category with the heading "Songs of Charos and Hades" (11).

In 1973 the autobiography of Markos Vamvakaris was published, prepared by A. Kail. She is distinguished compared to all the later biographies of other artists in two points. Firstly, the introduction is unbiased and reveals a good knowledge of history and social development of Greece. Secondly, Kail clarifies in detail how the text of the biography found its final form(12). For us however, most important is that, apart from a lot of songs that are reported in the text of the biography, Kail publishes an annexe of roughly 100 songs of Markos. It claims that he himself has transcribed all the songs from old disks and from tape recordings.

The same year (1973) T. Skorelis and M. Ekonomidis published a biography of Giorgos Rovertakis (GR) and the next year of Kostas Roukounas (KR)(13). And the two biographies contain an important selection of songs of the two composers. Most of them are accompanied by musical manuscript. It appears that all the material emanates from the personal files of the artists.

In 1975 Gail Holst published the book "Road to Rembetika" (I1). In this is included a small selection of Rebetiko songs that is accompanied by a translation in the English. It is supposed that Holst has written the songs from first publication of old disks of 78 rpm. This book gives a presentation of Rebetiko that is very personal and would be supposed it is only characterized as "appetizing". And Holst, of course, does not claim it represents a completed scientific study of Rebetiko.

If we exclude the many-paged second publication of Petropoulos's anthology and the anthology that accompanies Gauntlett's thesis, most recent is the "Rebetiko Anthology" of T. Skorelis (Sh.). This comprehensive work came out in four volumes 1977-78(?). The classification of songs here is based on the composers. Each section of the book begins with a short biography of the composer in question. Skorelis declares that his aim is firstly, to prove who has actually written the music of each song and second, to present the texts in their original form.

Today, however, we have at our disposal a sufficiently large choice of Rebetiko texts. But unfortunately all the anthologies that we reported are unreliable. A lot of songs have been published in two or more anthologies. And from them relatively many have come out in re-publication of old 78 rpm disks. Now we will analyse three such cases in order to show how problematic the situation is (Cf. the examples in heading 3.6).

Example A (p. 24): "My Painted Boat" by Giorgos Batis. HMV AO 2239 in the annex 3 list 1935. (The disk was recorded that is to say in 1935).

The variant of Petropoulos has one verse more than that of Skorelis. The remaining seven verses are identical, if we exclude the typographical error of Skorelis in the word "bouzouriera"(14). But when one hears the song in the re-publication, the situation becomes still more complex. Here the song is constituted from nine verses where the 6th, 8th and 9th are not included in the anthologies. Conversely, neither the first verse in Petropoulos nor the penultimate one in the two anthologies exist in the disk.

It is obvious that Skorelis and Petropoulos did not know the original disk. But unfortunately they do not reveal, as usual, the source of their variants. Moreover, Petropoulos dates this song in 1931 without giving other elements, while we know that the first disk with accompanying bouzouki was recorded in 1933 (see Section 4.1).

Example B (p. 25): "Kapnoloudes" by Bayanderas. Columbia DG 6106 in the general list 1936. (The disk was recorded in 1935).

The variant of Skorelis contains only three verses, while that of Petropoulos is constituted from four, as in the disk. Apart from two or three details the two first verses are the same, and in the three variants. But the two last verses in Petropoulos differ a lot from those on the disk, and are presented indeed in reverse order. Skorelis reports also in each verse variants of the last hemistich, again without revealing the origin. Moreover, he presents also a completely degraded variant, which was heard in a concert in 1972.

And in this case it appears that the two anthologists have not heard the original disk. But also the two know that the song has been recorded. Skorelis dates it, as is his habit without documentation, in 1935, and adds that it is the first disk of Bayanderas. And the two pieces of information are reliable. Petropoulos supposes that the disk came out in 1936, exceptionally a successful guess!

Example C (p. 26).

The problem here has a different character from the previous examples. It concerns two songs of Markos Vamvakaris, "Eprepe na 'rchosouna, manga mou" that according to him was the first song that he recorded and "O Dervisis" that we did not find unfortunately, only that in the anthology of A. Kail. Neither of these two songs could be found in the publications of old disks that we had at our disposal or in the lists of phonographic companies.

Gauntlett however has himself transcribed the first song from a tape recording of the original disk, and has confidence in his own transcription. And Skorelis, instead, presents two variants with the title "Pos ithela na 'rchosouna" (A, page 233). One of them is almost the same as that of Gauntlett. The text of the same song is found also in the autobiography of Markos (K p. 285). Here Markos, at the request of the publisher, quotes the song from memory. According to the preface (p. 9-10) it should date from 1969. Then Markos had forgotten two from the eight original verses. Moreover he had changed the order of the verses. Except from them, the text is almost same with that of Gauntlett.

Later, Kail in her anthology (p. 330) published a song titled "Dervisis". This of course, greatly resembles the other song, but does not contain a single identical verse. So far, so good.

Petropoulos however creates on his own a complete muddle with the "song" that he publishes, titled "Dervisi Mou" (P2 page 137 a' 1). His text is constituted from seven iambic fifteen-syllable verses. Such a form is unthinkable on a disk of that period, where the fifteen-syllable songs in sound recordings were only constituted from four or five verses. Admittedly, Petropoulos does not claim that he has heard his own variant on disk. In his enigmatic footnote however he refers to the anthology of Kail: Last three verses were published by Kail, he says. That is to say, Petropoulos copied from the anthology of Kail (p. 330). It is of course, identical in the two variants, with the unique difference that the last word "kataskefazo" (I manufacture) has been replaced in the Petropoulos with "paraskefazo" (I prepare) - that sounds still more strangely here that it is used for the versification.

Then, Petropoulos adds that these three last verses appear to belong in another song. Does no-one wonder, what made him then also include it here! Also one wonders, how it could escape his attention that two first verses of the song "Dervisi" in Kail, with insignificant variants and in reverse order, are also included in his own "song" as second and third verse respectively! Moreover it is perfectly mysterious how Petropoulos carefully avoids (or... that he omitted to mark) that the first and fourth verse of his own variant is the very same snippet with first and the third verse of the song that Markos himself mentions from memory in 1969 (K p. 285). It is indeed very curious, and Petropoulos refers already to the autobiography of Markos (K) saying that it is the first song that Markos put on disk. This information is found in page 159, in phraseology that obviously emanates from the initial autobiography of Markos himself. (Petropoulos, unlike Gauntlett, does not refer to the particular page).

Later from this complete trouble that he created himself, Petropoulos dares to accuse the anthology of Kail for "the frivolity with which it makes the classification of songs"! This criticism should in our opinion be reversed against himself, while the anthology of Kail compared to all the other anthologies is judged positively(15)

The three examples that we presented are certainly among the worst. But they are not unique - except perhaps for the last one. If two or more anthologies publish the same song, the texts differ almost always in bigger or smaller degree. In the few cases where two texts were identical, we were forced to suppose that one was simply a copy of the other. Next, Petropoulos (P2) alone refers to other anthologies, most times to that of Skorelis.

The unreliability of texts that have been published up to today constitutes a big problem. Still more deplorable is the fact that the editors almost always omit to refer to the particular source of each song. As we said already, Kail declares that she has transcribed all her songs from old 78 rpm disks and from tape recordings (p. 290)(16). Holst says that she transcribed most songs from published old disks. She does not reveal however the origin of the remaining songs (17).

Skorelis reports the chronology and the name of the singer of certain songs. It is supposed that this means that the song has been recorded and that the transcription has come from the disk. But, for the remainder (that constitutes the big majority) he does not reveal the particular source of each song.

Petropoulos, as we said, copied certain songs from Skorelis, referring of course to the source. Also he publishes certain songs and extracts of songs that he recorded himself while imprisoned under the junta. But in all the other cases he omits to refer to the source. Consequently we do not know if he himself (or some other) transcribed the song from old disks of 78 rpm or from later re-publication, or if he copied from authentic handwritten verses or from copy of such manuscript, or if he put verses that he mentions from memory somewhere in the 60's or the 70's, or if he mentions a live performance that he heard before, or...

3.4 The problem of chronology

The second basic problem concerns the chronology of the songs. Kail does not attempt to date the songs in her biography of Markos. Holst obviously will have based her dates on the "information" of Petropoulos (P1)(18). One cannot ignore, of course, the fact that the phonographic companies did not work on recording their assets (p. 53). But it is not obvious if one keeps in mind the censorship of Metaxas and its consequences.

In his introduction (p. 15-16) Skorelis stresses that the chronology of songs constitutes an important and difficult question. He is also more careful in this subject than Holst. Obviously he only reports the chronology of a song, if he has available certain elements. He is not satisfied with guesses. But according to those who are named on page 16 of the introduction, Skorelis sometimes supports the chronology with very doubtful sources. (He even attempts to date songs that have not been recorded on disk). He had at his disposal a general list of disks of Columbia for 1937, but it is not by any means obvious in what degree or how he used it. In Volume A (p. 15) he promises a complete bibliography, but unfortunately does not keep this promise. And the thought did not occur to him at all to present a discography!

Skorelis of course, keeps in mind the fact that the production of disks was stopped in Greece in 1941-1945 (Volume A, p. 16). But he does not appear to know that they did not produce Greek disks until 1917 in the USA (19) or up to 1925 roughly in Greece itself (20) (Volume A, p. 37-39). Also he wants to place the activity of G. Vidalis and A. Menemenlis in the USA, while actually they lived in Greece (21). Besides, he does not discuss by any means, what relation the American material has with an anthology of Greek Rebetiko songs. However, our basic accusation is that he almost never reports the elements with which he supports the chronology.

All these objections that we raised to the chronology of Skorelis, are still in effect in greater degree for Petropoulos. Again the bibliography and the discography are also absent. Also he reveals even more limited knowledge of production and circulation of disks etc. When e.g. he says that a song has been recorded, this does not necessarily mean that he has heard the disk. Then, in the cases where Petropoulos does not allocate elements for the chronology of a song, he inevitably often attempts a conjecture. Then, by common consent, he always puts a question mark behind the chronology. But this does not mean that the chronology is right in the cases where the question-mark is omitted, far from it!

In any case, we should admit that the big second publication of 1979 compared to the first has been improved a lot in this point. E.g. He shows more "moderation" precisely in the two cases where Ole L Smith comments (22). In the new publication he also avoids dating sound recording disks in the time interval 1941-45 generally and sound recordings of hashish related songs concretely in years 1937-40. But despite all that the new anthology of Petropoulos is still more unreliable than that of Skorelis.

Any student who does not keep in mind the unreliability of anthologies with regard to the output of texts of their songs and chronology is in danger of reaching perfectly absurd conclusions. The study of Damianakos (D.) is one of the best (or... rather, worst) examples. Even if equipped with an impressive theoretical training and with better intentions, Damianakos fails completely in the interpretation of the dependence of Rebetiko on social developments. He bases his analysis on the first published anthology of Petropoulos and accepts, without no discussion, the attribution of songs and their chronologies - even hypothetical ones! The result is that the conclusions to which Damianakos is led remain almost worthless (23).

3.5 The recent developments in Rebetology

The problems that we presented above have already been discussed by Gauntlett in Chapter 2 (p. 3-48). He is the first student to make minutely examined control of attribution of their songs and chronologies from the previous anthologies. Then, he refers each time to the sources of songs that he first published and reports also the elements on which he bases their chronology. He has himself transcribed a lot of songs from original disks and tape recordings. Moreover he is most assiduous in the transcription of songs and, like others, marks all the repetitions. Consequently, the texts that he first published are most precise (24). But also his approach, generally, in his study of Rebetiko and his strict scientific methodology constitutes big progress. Gauntlett has contributed considerably to the improvements in Rebetology.

In a "digression" (p. 195-200) Gauntlett makes a short critical examination of the development of Rebetology in the period 1978-1982 (that is to say the years that passed from the submission of his doctoral thesis up to its revision, its appearance, its publication). Even if somewhat superficial, this presentation is well-aimed.

We agree with Gauntlett in his assertion that the various biographies of Rebetiko artists that were published in this time interval do not contain only that minimal useful and sufficiently argued information. In our opinion the editors seldom show the faculty to appreciate rightly the value of the information that they transmit. Many times they appear incompetent to distinguish important verifiable information from gossip or obvious calumny. This is in effect in very big degree for Hatzidoulis who was the person in charge for most publications.

We also absolutely agree with the criticism of the book of D. Liatsou (25). It is incomprehensible that they accepted for publication such a book in 1982 and they remove already second publication in 1985.

Gauntlett simply reports, without particular comments, the publication of original 78 rpm disks that circulated during this period. These, in our opinion, constitute without equal the most important contribution to Rebetology. We should recognise the energetic activity of Hatzidoulis precisely in this point. Gauntlett consulted the few publications that had come out before he prepared his publication. To us it appears strange that he does not stress the big value of these publications. From 1982 up to today they have published even more old disks than in the previous period. Consequently we today have at our disposal sufficient material. (For publication see below Chap. 4.3).

3.6 Examples

Example A

Example B

Example C

4 Records

4.1 Sound recordings and production

The first known Greek sound recording disks were made in Smyrna and in Constantinople in the second decade of our century. In the available publications (26) we did not find any song on the subject of hashish or narcotics, consequently there does not exist any thematic relation with our own songs. But musically, the relation is obvious. The Asia Minor songs were recorded with one or more from the following musical instruments: violin, santouri, kanonaki, outi, lyra, accordion, banjo and clarinet. The music is based on particular scales, that are called "makamat" in Turkish and "dromoi" (roads) in the Greek, often used dancing rhythms and the songs begin with a instrumental improvisation, called a taximi (27). The song is sung in a soprano or tenor voice, called "Katsares voices" (28). This musical delivery is named Smyrneiko style or Café-Aman, from the cafés where they played "Turkish music". The existence of such cafés is already also argued in Athens from the end of the previous century (29). After the Catastrophe of 1922 the delivery was strengthened in mainland Greece with the arrival of many Asia Minor composers, musicians and singers, amongst whom we can report Evangelos Papazoglou, Spiros Peristeris and Evangelos Sofroniou. Rebetika that were recorded on disk in Greece up to 1933 all belong in this delivery (30).

In the USA disks in the Café-Aman style had already been recorded from about 1917. Marika Papagika began her career there in 1918 and the next 20 years is presented in more than 200 disks (31). She also recorded on disk various hasiklidika songs, e.g. certain variants of "Baglamades" (32), and one of them is found with four other disks of Papagika in the list of Greek Columbia 1934. These disks were imported from the USA and circulated in Greece (33).

Around 1925 (34) the English "The Gramophone Company" (HMV) began, in two improvised rooms in Athens, the sound recording of disks in Greece. Sound recordings were made in "wax", that is in disks made from mineral wax, that they took to England for the final production of the disks, which later were exported to Greece (35). At the end of the twenties, HMV, Odeon, Parlophone and Columbia offered disks that had been recorded in Athens for sale in Greece (36). Each company employed as maestro a musician whose work was to choose suitable material, engage players and supervise the musical side of the sound recording (37). The sources that we have at our disposal give contradictory information about who worked for which company. They agree however in that the maestro in Columbia was P Toundas(38) who proved to be one of the most fertile composers in style of Café-Aman. It was he who prompted Roukounas to record his first disks at the end of 1928 (39). At HMV the maestro was D. Semsis (Salonikios) who was considered to be the best violinist in the Balkans and played accompaniment on countless sound recordings (40).

In 1930 the disks were by now so much coveted in Greece that Columbia decided to build a disk factory in Athens, which was used by all the companies that manufactured disks (41).

At the end of the twenties they began in the USA to record songs with a new style: with solo guitar or bouzouki and the previously mentioned "Pireotiko voice" that can be described as a monotonous baritone (42). With guitar accompaniment is presented G. Katsaros that recorded hasiklidika songs, and K. Dousas. Only a few disks with bouzouki were recorded in the beginning in the USA. One of first is with Ioannis Halikias (song) and M. Karapiperis (bouzouki), recorded in 1929 (43). A little later Ioannis Halikias recorded with bouzouki (with guitar accompaniment by S. Michelidis) the "Minore Tou Teke" and "The Mystery", that made a huge impression on Greek musicians and maestros when the disk reached Greece in about 1932 (44). It appears that this recording was printed by Greek Columbia two times in the thirties, with numbers DGX 36 and DG 275 (45). The first reprint is found in the list for 1940, not however in preceding lists - the last one does not exist in any of our lists, it should however be dated before 1934 (46). Halikias' disk is the only American sound recording for which we have personal testimonies of Greek musicians and that appear to have influenced regular sound recordings in Greece (47). Papaioannou, for example, writes that it drove him wild as soon as he heard it - he had heard other disks than the USA ones that had not made an impression on him because they were monotonous (48). And Roukounas describes the big impression that the disk made in musical circles and how Peristeris put Zaharia Kasimatis to play Halikias' piece on the guitar - "however something of a sad song, being him, said Zaharias". Roukounas reports also the "Kalogeraki" of Douss which reached Greece a little after Halikias' disk (49). Two disks by Dousas were printed by Greek Columbia with numbers DGX 28 and DGX 38, that are found in the lists for 1934, 1936 and 1939, and 1936 and 1937 respectively (50).

The first to bring out disks with bouzouki in Greece was Markos Vamvakaris (51). From himself and from other contemporary musicians we have contradictory information for when precisely it was made, also in whose company (52). Markos himself claimed that his first sound recordings were hasiklidika songs: "Eprepe na' rhousouna" (B 6) and "Harmanis eim' ap' to proi" (B 26) (53). His oldest songs however that we could date with certainty in the available material are "Efoumernam' ena vrady"(B 7) and "Tragiaskes" (B 25) that are found in the general list of Columbia 1934 with number DG 326 and in that of Parlophone 1934 with number V21710 respectively. Consequently we can consider as given the chronology of 1933 for the first sound recordings with bouzouki in Greece. The fact that recorded disks with bouzouki and baglama were first made in Greece, is owed to the infamy of two instruments that were connected closely with hashish usage and prisons (54).

A little after the first appearance of Markos on disk, the companies engaged also other bouzouki players. The most famous that recorded disks between 1933 and 1936 were, apart from Markos, Anestis Delias, Giorgos Batis, Dimitris Gogos (Bagianteras), Stelios Keromitis, Yannis Papaioannou and Mihalis Genitsaris (55). These seven musicians who constitute the Pireotiko school played exclusively bouzouki and baglama and sang with nasal, throaty voices. There are however certain common points between the style of Café-Aman and the Peiraiotiko style with regard to each other musically: they use the same roads (dromoi) and dancing rythms and also taximi (56).

There exist however elements that prove that the musicians of Café-Aman simultaneously began to record disks with bouzouki. Spyros Peristeris plays bouzouki in "O Teketzis" (A 31) that is found amongst the new disks in the list of Parlophone of January 1934 (No B 21707) (57). Also recorded were disks with Giovan Tsaous (Yannis Etsireidis) who played saz, the sound of which very much resembles that of the old bouzouki (58).

Gauntlett claims (p. 90) that the popularity of music from the school of Café-Aman was in danger after about 1934, and that the period 1932-1940, that he calls "classic", took over from Markos and the other bouzouki players of the "marginal life" (p. 51). With regard to this we should however underline that the lists up to 1937 included, present an unmistakable predominance of disks in style of Café-Aman. The lists offer also another clue to the calculated commercial dynamic of certain songs. Certain songs on the subject of hash and narcotics circulated almost simultaneously from more companies, e.g. "Pasalimaniotissa" of Toundas, (A 33) where his two implementations, with G. Vidalis and M. Politissa (33 a), are reported in the list of Odeon for 1931 and a third with "Arapaki" (33 v) in that of Columbia for 1934 (59). Also we present "Ferte Preza" (A 41) in the list of Parlophone of January 1934 with Stella Bogiatzi and in the list of HMV 1935 with Rita Abatzi. These elements however show only the availability, and as we do not know what circulation the disks had (60), we cannot immediately calculate the relative popularity of the styles of Café-Aman and Pireotiko for this period.

4.2 Censorship

After the imposition of his dictatorship Metaxas created a Committee of Censorship, according to A.N. 45/1936 of (29/31 August). Inspiration for this was Theol. Nikoloudis, amongst others in the Ministry of Press and Tourism (61). The censorship had as consequence an expurgation of thematic material of songs so that the songs on the subject of hash and narcotics were also removed from the lists.

Gauntlett writes (p. 340) that the last sound recording of Hasiklidika songs of HMV had number LO 2220. However in annex 7 to the catalogue of 1935 we found "Ta Hanoumakia" (A 39) with number AO 2279 and "Xtes to vrathi sto skotathi (Yesterday evening in the dark)" (B 27) with number AO 2280(62). Gauntlett reports also that the last disk with a song of "marginal life" that Columbia circulated had the number DG 6192, but in the list of 1937 we also found "Kouventes sti filaki(Conversations in prison)" (A 12) that has number DG 6217.

We can with enough certainty claim that no new songs on the subject of hashish and narcotics were recorded after 1936. There continued however the circulation of certain old disks of this type: in the list of 1939 HMV we find e.g. "Hanoumakia" and "Htes (the evening)" with the same numbers as in the list of 1935, and "Prezakias" (A 27) with number AO 2295 and "Ego thelo prinkipessa (I want a princess)" (A 6) with number AO 2319. As a consequence of censorship certain songs came out in new variants, cleansed of references to hashish, e.g. "Ego thelo prinkipessa " by Kyriazopoulo (published on "Megaloi" 13), and "Conversation with Charos" (published by "Stellakis").

In the lists of HMV the songs for us are found in the department with the heading "Popular, Rebetiko songs, manedes, anatolitika" while for those of Columbia the heading is "Laika - Rebetiko songs". It is noticeable that the heading in the lists of 1939 (HMV) and 1940 (Columbia) has been changed to "Laika" - the term "Rebetiko" genuinely appears to have been connected in such a degree with the "underworld" that it has been "cleansed".

4.3 Republications

From about 1975 the phonographic companies began reissuing old 78 rpm disks transported to disks of 33 rpm. From the technical opinion the quality of certain disks is so awful that the verses are almost inaudible, a thing that is in effect e.g. for certain of "Apagorevmena". The majority of reissues do not offer any information about the original sound recordings, such as the names of musicians and the chronology, or the matrix numbers and list that could contribute to the chronology. They do not even print the verses of the songs.

From the disks we consulted, only the recordings "Greek Oriental", "Papazoglou", "Giovan Tsaous" and "Delias" contain on the covers the verses, and the transcriptions in the first two at least correspond almost perfectly to ours. On the covers of disks "Papazoglou", "Giovan Tsaous" and "Delias" are entered also the elements printed on the labels of the 78 rpm disks, namely the title, the type of dance, the players and the instrumentation. They are reported, without source, and with chronologies that however do not correspond precisely with ours. The editor of "Greek Oriental" is sufficiently circumspect in the chronology of the sound recording and writes "Roughly...". He reports the names of players and the instrumentation, and this disk is also distinguished in that he is unique in recording the matrix numbers and list.

Generally the name of the composer is reported in republications. A striking exception of this rule is "Apagorevmena (Prohibited)", the editors of that appear to fear the charge for violation of copyright - so we do not say that it is clean theft of rights. Certain songs here are attributed to Marika Papagika, when we know she was not the composer. In the case of other songs it is only reported that the piece is an arrangement by Skarvelis or Kyriazopoulos, but all that is known is the name of the real composer.


5.1 Introduction

We used almost completely the system of transcription of Gauntlett(p. 202) with the following symbols:

X 2 (X 3) after a verse:
The verse is repeated a time (two times).

X 2 after one couplet, in between in the two verses:
Entire couplet is repeated.

2 - 1 after one couplet, in between in the two verses:
The verses are repeated in reverse order.

Words in parenthesis in a verse that is repeated:
The first time the verse is heard as it is written, but in the repetition 
the contents of the parenthesis are replaced with the contents of the 
parenthesis after the verse. The number in this parenthesis shows the 
repetition, in what becomes the replacement.
(-) :  The word is omitted.
< > :  Doubtful transcription.
....:  Words that are not heard.

Unlike Gauntlett we did not try to make a hypothetical completion of the verses, we do not mark the anti-prosody tones. We copied the songs of Gauntlett with the monotonic system and wrote the words with incidental phonological variants (e.g. nargele in A 1). Moreover we put in parenthesis phonemes where, while we do not hear them, they are essential in order to give the verse meaning.

The division into stanzas was based on the melody, to the extent allowed by our deficient knowledge of music, and not obligatorily based on instrumental fills, because these in certain songs are found in the middle of a couplet (e.g. in A 1 and 25 in each stanza, and in A 2 and 11 in every second stanza). There also exist songs that do not have fills before the refrain (A 10 and 27).

Moreover we record interjections that do not constitute part of a song at all, by which we often can ascertain the identity of the singer, and in certain cases also of the instrumentalist. They are preceded by a hyphen and have an exclamation mark after them.

The titles of our songs emanate in most cases from the lists with the disks. For the songs that we did not find in the lists we chose the title of the republication (63).

The elements for each song are categorised as follows:

  1. Source: Disk/Gauntlett.
  2. Singer
  3. Composer. Author of verses.
  4. Instrumentation.
  5. Basic rhythm.
  6. Type of dance.
  7. Chronology: Phonographic company, number of disk, [chronology of sound recording], chronology of list. (Where there does not exist any other clue, it is the general list).
  8. Anthologies k.a. that present a variant of the song.
  9. Other titles.

The elements emanate above all from the catalogues, that include all the disks of the company, both new and old, that were offered for sale at the time in question. The new disks were numbered in chronological order and reported the title of song, the dancing rhythm, the composer and the singer and sometimes also a musician. The musical instruments however are only marked when it is the bouzouki and baglama. The author of the verses is not reported, because the composer usually also wrote the verses (64). Amongst our songs there are only a few exceptions to this rule: "Mas kinigoun ton argile" (A 16) by Lorentzou, the verses were written by Nikos Mathesis and "M' argile kai baglamades" (A 14) with verses written by Stratos Perpiniadis. Vosporinou (A 3) and Delta (A 7) are pseudonyms. For the songs of Tsaous it is in effect generally that his wife wrote the verses (65).

For only one of our songs do we have the date that it was recorded (A 24). We based the chronology of the remaining songs on the catalogues that give one sure "terminus ante quem" (66). The general catalogues came out at the beginning of the year, and the new disks that they included had already been recorded in the previous year. We report this in parenthesis and other versions of songs in question. For the songs that we did not find in the catalogues the indeterminate "terminus ante quem" is in effect, that is the end of 1936.

Apart from the catalogues we only had at our disposal one document that helped in the chronology of our songs. It is a receipt from the third quarter of 1934 that proves that that year Papazoglou took percentages inter alia for "Argiles" (A 2) and "Marika Hasiklou" (A 15)(67).

The abbreviations that we use for republications are found in the discography (p. 149) and those for the anthologies, the biographies etc. in the bibliography (p. 158-151). We are reported however in the B publication of Petropoulos only with "P.", that is P. p. 142 B = Petro'poylos, B edk., page 142, Third column.

The information on the instrumentation (4), basic rhythm (5) and the type of dance (6) are owed, as we reported in the preface, to Lisbet Torp. With regard to this information we should underline certain problems that concern the songs of Café-Aman: it is very difficult for anyone to distinguish the guitar from the outi(oud) when they play accompaniments - it is equally difficult to distinguish the old bouzouki from the saz and guitar when they play the melody, because old bouzoukis had strings made from gut, and not from metal. With regard to the type of dance is also in effect for the two faculties that hasapiko is slow and that it is complicated for anyone to distinguish kamilieriko from heavy (syriano) zeimbekiko.

Songs A. Songs X. Songs B.

(1) In only one case we had at our disposal an old disk of 78 rpm belonging to Lisbet Torp (A 21 v).

(2) A 9, 14, 35, 40, X 1, 2 and B 3, 6, 26.

(3) According to Ian Miller traditional song H I was recorded with acoustic sound reception, that is to say before 1925 (G. A 85).

(4) Our translation of the term "low life".

(5) Certain articles that were published in the frames of this discussion are printed with other more recent in the annex of Greek publication of book of Gail Holst (H2).

(6) A. Xenos in the "Letter in the Rizospasti", 4 February 1947 (I2. p. 141).

(7) Px. F. Anogianakis in "Rizospasti", 28 January 1947 (H2 p 139 - 141)

(8) Px. F. Anogianakis and M Hatzidakis (H2 p 151 - 155).

(9) Problematically, there are a lot of comments above in the songs of the anthology such as "The song reflects old grumbling" and other equivalents.

(10) For example the song "Ymnos youth", P2 s. 213 b I. Besides, it should be remarked here that the Petropoulos does not give an explicit definition of Rebetiko.

(11) P2 page 160 a 1 where entitled "Farewell to Hades". See also on the same page "If I die" (160 g 1) that does not mention either Charos or Hades.

(12) Corresponding information is completely absent from the biographies prepared by Hatzidoulis.

(13) And in the two cases they claim that the text of book contains precisely what the artist said. But in the biography of Roukounas it is obvious that a part of text is only constituted by his own reasons. And in the case of Rovertakis the final form of text is owed certainly in the editors.

(14) The error is corrected on p. 348.

(15) We discovered only two serious errors in the anthology of Kail. In s. 322 a song is published titled "Otan me vlepeis na perno" that is constituted from seven fifteen-syllable verses. But as we said already, the songs of this metre were constituted from four or five verses in sound recordings of the period. And because we recognize the last four verses as the complete song "Yia to ginati sou mori" (noym. B 5), we draw the conclusion that here two different songs e'hoynenwcej' at error. In s. 328 are only published with the title "Blokos" two first verses (from the total of four) the song "Efoumernam ena vrady" (B 7). We suppose that the Kail transcribed these two songs from bad quality tape recordings.

(16) Kail is also the first author who stresses how very difficult it often is to hear what precisely is being sung in the old disks. She has already expressed her general reserve for the right attribution of songs of (p. 29c).

(17) Certain of the songs in I1 are not recorded on disk - at least not initially. This is in effect e.g. for the "Litany" of Tsitsanis (p. 110) that first came out in disk in 1982 (Venus S. V. 75).

(18) E.g. I1 p. 97 where the song "O Oropos" of Batis according to P1 is dated in 1933. But the disk came out in 1935. It is found in annex 3 of the catalogue of HMV for 1935 with number AO 2239 and title "The prisons of Oropos".

(19) See the notes of D. Spottswood in his re-publication "The Greek Popular Song in America" (AF 22/23). See also OS1 p. 178-79 and c.f. the criticism of re-publications "The First Rebetiko Songs 1900-1913" that Skorelis took care of for the Greek CBS with number 53753.

(20) See also OS1 p. 184.

(21) In the Greek lists of disks that we had at our disposal the name of Vidali is presented often, also Menemenli few times. It is obvious that they took part in recording disks in Greece. See also. chapter 4.1, heading 30.

(22) 1) OS1, p. 182-3 for "Manolis the hashish smoker" of Dragatsi: "P1 reports: Disk of 1920(?)", while P2 s. 127 g 2 he writes: "Hasapiko of Dragatsi. In 1926 recorded by Nouros...". Petropoulos namely anymore it has passed the magic limit 1925. But it appears that it did not advance enough, because the disk with Nouros is reported in the general list of Columbia for 1934 with number 18079. (The disks of line Col. 18000 were printed in England and because of this chronology for them are particularly difficult. Organisation of Producers. Kounadis says that it is dated in 1930 - without however offering references ("Mousiki", Dek. 1980, 26)).

2) OS1 p. 184 for the 'Heroin and Mavraki' of "Gavala: "P1 p. 276 he remarks: Disk of 1922(?)", while in P2 p. 134 g 3 he writes: "Disk of 1925(?) with Stellaki... It is supposed that the composer is Gavalas...". Again Petropoulos reveals that he has not exceeded his absurd preference for precocious chronologies. The variant with Stellaki is reported in the list of Columbia for 1936 with number DG 6126. (Gauntlett who had this disk in his hands reports it as number DG 6162, (p. 257)).

(23) For a detailed criticism of Damianakos see G. p. 18-21 and passim.

(24) But also the transcription of Gauntlett can in certain cases become still more precise. Compare e.g. his song with number C 24 with our own transcription "Mes stou Zampikou" (A 19).

(25) "The refugees of Asia Minor and the Rebetiko song" by Dimitri Liatsou, Athens, 2nd edition 1985. (See G. p. 195-96, para. 2).

(26) The 33 rpm disks "Greek-Oriental" and AF 14 ("Authentic songs recorded in Smyrna and in the City before 1922").

(27) "Greek-Oriental". Dietrich s. 110 and 131 k.e. - Of the instruments the clarinet is not only used in Rebetika.

(28) G. p. 68, Dielrich s. 14.

(29) G. p. 65 k.e., Mazaraki p. 48-50. See also Thodoros Hatzipantazi's "Asiatidos Mousis erastai'", moment, Athens 1986.

(30) L. Menemenlis and G. Vidalis who sang in sound recordings in Smirna around 1915 are presented later in disks recorded in Greece after 1922. Two by them are included in our anthology (H 1 and H 5). See. ch. 3.4, par. 21

(31) Spottswood

(32) G A 77 and A 82

(33) From the end of the twenties the American companies imported moulds from Greece and brought out disks in the USA with Roza Eskenazi, Rita Abatzi, Papasideri, Dalga and others (Spottswood and "Authentic" 67 and 70).

(34) In 1925 electric sound reception replaced the acoustics. Because the method of sound reception is reported in the label of disks, it can contribute to the chronology of recordings. (G.p. 86, par. 133)

(35) KRs. 18.0. p.74.

(36) G. p.74, par.84.

(37) G. p.74.

(38) KR p. 17, 29 and 30. GR p. 15. Pap. p. 39. "Ta Tsilika".

(39) KR p. 17 and 20.

(40) The list of HMV 1935 and annexes "Greek-Oriental".

(41) G. p.75, par. 89.

(42) G p. 86-87 and mark 133.

(43) OS2 and Spottswood. The songs are found in A 44 and 45.

(44) G p. 87 and Spottswood.

(45) G p. 87, mark 136. The information emanates from KR, p. 30.

(46) Our song A 19 is found in the list of Columbia 1934 with number DG 273.

(47) On 20 May 1930 A Kostis recorded with guitar at least four Rebetiko songs for HMV in Athens. It appears however that it remained in the USA and did not have any relation with musicians in Greece. Moreover sound recordings they were intended for circulation in the USA and do not exist in any of our lists. (G. p. 88. Vl. G. A 20 a and A 46).

(48) Pap. p. 21.

(49) Kr p. 30 etc

(50) G p. 87 mark 136 it only reports DGX 28.

(51) The first that recorded a disk with bouzouki was Batis, in 1933, a few months before the disk of Markos was circulated (X. p. 22-23 and OS2, note. 8).

(52) G p. 88, note 140 and Pap. p. 39.

(53) G p. 88 - information is based on K p. 158 etc

(54) G p. 86, note 134.

(55) G p. 89-90. It should be noted that the three last musicians recorded hasiklidika Rebetiko songs in this period.

(56) "Greek-Oriental".

(57) Peristeris also plays bouzouki in A 16 and 20. Vl. and Sh. G, s. 142.

(58) A 6, 27, 32, 34. Vl. 0. p. 90, note 148 - Indeed, and Gauntlett and the lists write that Tsaous plays bouzouki. For the sound of musical instruments see below, chap. 5.1

(59) This song exists also in a fourth implementation with the A. Dalgas published on "Megaloi" 13 (A 33 c).

(60) Only that for certain disks of Papazoglou (Sh. G, p. 78. Vl. and below ch. 5.1).

(61) See. "Rebetiko songs of Possession", Month MSM 391, assiduity K. Hatzidouli.

(62) G. had at his disposal only the general list 1935 (p. 366).

(63) With the appearance of the republished "Delias" were revealed the initial titles of songs B 4, 14 and 18. Songs A 5 and 28 also have title errors.

(64) Sk passim KR p 49

(65) Sk B, p 46 re-issue "Giovan Tsaous".

(66) Unfortunately certain of the handwritten extracts of catalogues that we had at our disposal were undated. Also we do not know in which month the annexes of the HMV catalogue came out.

(67) Sk C, p 78.

Songs A. Songs X. Songs B.