Existing Hypotheses about the Emergence of Nonsense Syllables in the Chant Tradition of Teretismata and Kratēmata in Byzantine Music

In the present essay, I review and evaluate the three main hypotheses about the historical background and emergence of the nonsense syllables in the chant tradition of teretismata and kratēmata in Byzantine music. The different historical hypotheses as to the historical roots and development of this singing practice are examined and analyzed thoroughly, namely those of Gregorios Stathis (1979, 2014), Diane Touliatos (1989), and Grigorios Anastasiou (2005).
The aim of the analysis is to summarise and discuss the contribution of the up-to-date historical hypotheses to the theoretical approaches of the topic, including the identification of potential flaws, lacunae and inadequacies of their explanatory power.
Touliatos takes antiquity as a starting point for her hypothesis and posits the roots of nonsense syllables in the music of Ancient Greece. After a historical gap of several centuries, those syllables reappear in the Byzantine music during the 14th c. Stathis and Anastasiou examine the phenomenon exclusively within the boundaries of Byzantine music. Before reviewing the three hypotheses concerning the historical development of nonsense syllables in Byzantine music, we must consider the sources on which the theories are postulated.

I n the present essay, I review and evaluate the three main hypotheses about the historical background and emergence of nonsense syllables in the chant tradition of teretismata and kratēmata in Byzantine music. The different historical hypotheses as to the historical roots and development of this singing practice, namely those of Gregorios Stathis (1979Stathis ( , 2014 1 , Diane Touliatos (1989) 2 , and Grigorios Anastasiou (2005) 3 are examined and analysed thoroughly.
The aim of the analysis is to summarize and discuss the contribution of up-to-date historical hypotheses to theoretical approaches of the topic, including the identification of potential flaws, lacunae and inadequacies of their explanatory power.
Touliatos takes antiquity as a starting point for her hypothesis and posits the roots of nonsense syllables in the music of Ancient Greece. After a historical gap of several centuries, those syllables reappear in the Byzantine music during the 14 th century. 4
Just like kratēmata, teretismata evolved to become independent musical compositions of the repertoire of Byzantine music. 15 The 14 th century is characterized by the appearance of revered individual composers in Byzantine music. According to John Plemmenos, during this period, the kratēmata reached their artistic peak, but after the fall of Constantinople to the Turks in 1453, the musical production of kratēmata stagnated until the late 16 th century when it was revived again. 16 The use of nonsense syllables in Byzantine chant is not limited to teretismata and kratēmata. Before starting to sing in each mode, the domestikos (precentor) gets properly "tuned", so to speak, in order to introduce the ēchos (mode) by singing a melodic intonation formula consisting of nonsense words. This intonation formula is called ēchēma, epēchēma, or apihima and each ēchos is allocated a name of the mode (see Table 1).
According to Werner, the well-known formulae from the Byzantine theory and from the writings of Aurelian of Réômen in the medieval Western church, neannoe, noeagis, noeagis, etc., have caused many speculations about their origin and their appearance in Latin psalms. 17  According to Conomos, although the Western formulae were imported from the Byzantine East, they differ from them in that context, textually and functionally. 19 Touliatos's hypothesis: incantate, solmization and glossolalia function of the teretismata and kratēmata Diane Touliatos published in 1989 probably the most detailed review article about nonsense syllables in the music of Ancient Greek and Byzantium. 20 Touliatos points out that the appearance of nonsense syllables in the music of ancient Greece and Byzantium can be traced to the use of the seven Greek vowels in gnostic music. The nonsense syllables in Ancient Greek music functioned as incantations and were linked to the seven vowels α, ε, ι, η, ο, υ, ω of the Greek alphabet in gnostic music, something that, according to Touliatos, is well documented in many historical works and treatises: The appearance of nonsense syllables in the music of ancient Greece and Byzantium can be traced to the use of the seven Greek vowels in gnostic music. From antiquity through the medieval period, the vowels α, ε, η, ο, ω, are discussed in many historical works and treatises for their function as incantations. It is certain that these gnostic formulae were in existence long before they were documented. 21 One of the first treatises referring to the aforementioned role of nonsense syllables is the Handbook of Harmonics written by Nichomachus of Gerasa in 2nd century A.D. Touliatos anchors her interpretation of the role of the seven vowels of the alphabet, the incantantory function of nonsense syllables, and the number seven by referring to the Pythagorean treatise of Nichomachus. According to him, as Touliatos mentions, seven was considered to be a magical number, as each one of the seven spheres produced a sound corresponding to each one of the seven Ionian vowels. A for the Moon, E for Mercury, H for Venus, I for the Sun, O for Mars, U for Jupiter, and Ω for Saturn. The same claim about the mystic affiliation of the seven vowels with the seven planets, is also found in another treatise, The Elocution, written by a contemporary of Nichomachus, Demetrius Phalereus. There, it is documented that the Egyptian priests "worshipped their gods by chanting the seven vowels which designated sounds or pitches and which were substituted for the performance of the aulos or kithara". 22 According to Touliatos, the gnostic formulae of the seven vowels corresponded to the musical notes of a heptachord, and thus a new musical system was created. 23 These gnostic formulae of the seven vowels functioned as invocations sung totally as nonsense, with rapid movement of the notes that caused a trembling of the voice. According to Touliatos, the definition Psalmodic Formula "Neannoe" and Its Origin". 19 Conomos, "Ēchēma". 20 Touliatos, "Nonsense Syllables," 231-243. 21 Touliatos, "Nonsense Syllables," 231. 22 Touliatos, "Nonsense Syllables," 232. 23 Touliatos, "Nonsense Syllables," 233. of this coloratura-like type of trill singing by the term teretism (multiple repetitions of same pitch-note corresponding to a certain vowel), first appears to the treatise On Music, written by an anonymous Hellenistic author. 24 For the explanation and documentation of the addition of consonants together with vowels during the evolution of the nonsense syllables, Touliatos refers to two theoretical treatises from antiquity, About Music by Aristides Quintilianus (written between the first and fourth century A.D.) and Bellermann's Anonymous (of uncertain date, first published by Fr Bellermann in 1841 in Berlin). 25 Quintilianus, who has had a wide reception in the study of music and rhetoric in Europe, talks about the seven vowels although he maintains (still according to Touliatos), that only four of them, corresponding to the musical notes of a tetrachord, had the desired sound qualities and were appropriate for solmization: alpha (Α), epsilon (E), eta (H), and omega (Ω). The same author reports that the most appropriate consonant to juxtapose with the vowels was tau (T), the consonant that sounds like a string instrument and precedes the vowels in all Greek articles. The selection of the consonant tau created the combinations of syllables τα, τε, τη, and τω. The latter were used in solmization practice in Ancient Greek music and assigned to the different pitches of the tetrachords. Touliatos mentions that the Anonymous treatise presents -with few differences -the same facts about the solmization music system in Ancient Greece which was based on the four vowels and the consonant tau used for singing the musical intervals. Bellermann points out that this type of solmization system created three types of articulation, kompismos, melismos and teretismos. 26 Touliatos indicates that, between the period of antiquity and the 14th century, there is a theoretical gap owing to the lack of treatises. These nonsense syllables are discussed again in the treatise of Manuel Bryennios, the Harmonics, written in about 1300 A.D., which is influenced by Quintilianus, something that is obvious from the many affinities between the two treatises.
The tradition of nonsense syllables reappears in the Kalophonic melodic style of Byzantine chant during the 14 th century and is called teretismata. These begin with the consonant tau or rau and their function is to erase the gaping sound caused by the prolonged vowels. Teretismata were developed to the new kalophonic chant evolved further to independent and freely composed melismatic compositions based on nonsense syllables, mainly τε, ρε, ρο, τι, ρι, etc. The teretismata, when found in the concluding sections of the Akolouthia manuscripts, were called kratēmata 24 In Touliatos, "Nonsense Syllables," 234: "The treatise was published by A.-J.-H. Vincent, "Notices sur trois manuscrits grecs relatif a la musique," Notices et extraits des manuscrits de  and/or anagrammatismoi (anagrams), arranged according to the modes. Apart from the consonant tau and rau, other letters such as χ, ου, and γγ were in use. Although Touliatos supports the evolution of the kratēmata from the teretismata, she does not offer a convincing theory or hypothesis about how this transformation took place. 27 In many liturgical texts from 14 th , 15 th and 16 th centuries and even from the neo-Byzantine era, interpolated sections with compositions based exclusively on nonsense syllables were written by composers as a way for chanters to express themselves and show their mastery and ability. Touliatos evokes the hypothesis in the Exēgēsis of Gerasimos, a seventeenth century Cretan monk, concerning glossolalia, (the "wordless jubilation", thus the attempt to mimick the singing of the angels). According to Touliatos, the glossolalia is a possible explanation for the allowance and evolution of this type of singing in the strictly religious Byzantine empire, despite its roots back in the magic papyri and pagan rituals of antiquity. 28 Gerasimos explains how the teretismata compare to the running of rivers, the singing of birds and the trilling of cicadas, while the kratēmata are described as river and nightingale, or musical instruments as trumpet and bell. For the latter explanation, Touliatos refers to the kratēma entitled "A Bell", written by Gregoritze Domestikos in 1453 for the fall of Byzantine empire. In this kratēma, the vocalization of the nonsense syllables imitates, by the use of intervals of fifths, the chiming of a bell. 29 Touliatos concludes that this solmization practice, both in antiquity and Byzantine traditions, has in common syllables articulated as tereritism, a type of vocal ornament. Touliatos ends by claiming that this solmization system functioned as a mnemonic technique, which was further developed into the noeane formulae of the Western mediaeval music theory: It is not accidental that this solmization practice was found in both traditions but is evidence of Greek theory influencing Byzantine theory and practice. This ancient system of solmization not only influenced the Byzantine tradition but also provided a link to the mnemonic solmization practices which developed in the West, for it is probably from similar syllables that the noeane formulae of Western medieval theory were derived. 30 27 Touliatos, "Nonsense Syllables," 239-240. 28 Touliatos, "Nonsense Syllables," 240-241. 29 Touliatos, "Nonsense Syllables," 241. 30 Touliatos, "Nonsense Syllables," 243.
Stathis's hypothesis: affiliation of teretismata and kratēmata with ēchēmata in ecclesiastical music/the origin of the kratēmata by the ēchēmata A different theory about the history of Byzantine nonsense syllables has been proposed by Grigorios Stathis. 31 The ēchēmata as melodic intonation formulas constitute a distinct chapter in the theory of the Papadikai or other various theoretical books dealing with the theory of the ecclesiastical Byzantine music. The words of the intonation formula of each ēchēma corresponding to each ēchos (mode) of the Octoēchos, are presented in the following Table 1:   31 Primarily, the book used for research about the formulation of Stathis's approach is the English translation (see Stathis Touliatos (1989)   Apart from Hagia and Ne(h)agie (meaning Holy/Saint and yes-holy/saint), the ēchēmata lack conceptual context and appear to be totally nonsensical. It is generally believed that they made as little sense to a Greek speaker in Byzantium. For some conservative Byzantine scholars, the fact that music genre aiming to serve God lacks real linguistic meaning, is a scandal. 32 The ēchēmata are found in manuscripts dated to the tenth century AD, although scattered testimonials about the use of ēchēmata in secular and ecclesiastical ceremonies are found the Book of Ceremonies written by the Byzantine emperor Constantine Porphyrogennetos (b. 905-d. 959 AD). 33 If we take into account recent findings about the relation of ēchēmata and the intonation psaltic formulae neannoe, then the appearance of the former can be testified even earlier, almost simultaneously with the birth of the Octoēchos. 34 The ēchēma is recited by the domestikos in order to introduce the choir to the ēchos and its sound. Apart from this intonation's practical significance, the ēchēmata gradually developed into distinct ornamentation element, or kallōpismos, of the structure of kalophonic melismatic compositions. 35 Stathis states that the three main features of kallōpismos (ornamentation in kalophōnia are a) the kalophonic melos, which is more elaborate and eloquent compared to the papadic melos, b) the anagrammatismoi of the poetic text, and c) the ēchēmata or kratēmata. 36 The same scholar claims that the kratēmata originate from the ēchēmata and he uses the terms nenanismata, teretismata and teretismos as synonyms of the kratēmata. The kratēmata are mainly based on the nonsense syllables Tititi, Tiriri, Terere, Tetete and Terirem, while the ēchēmata on the nonsense syllables Anane, Anena and Tenena. The nonsense syllables of the ēchēmata establish the genesis of the kratēmata: 32 Anastasiou The designation ēchēmata, which was originally used for all the kratēmata, refers specifically to the kratēmata using the syllables Anane, Anena and Tenena. This particular instance is interesting because we can acurratelly establish the genesis of the kratēmata through these very syllables. 37 Although this affiliation seems to be valid, the hypothesis proposed by Stathis needs further confirmative investigation in order to explain sufficiently the transition from the syllables of the ēchēmata to those of the kratēmata. The nonsense syllables in the kratēmata are mainly linked to the consonants tau and rau, and thus consonants very different from the syllables of the ēchēmata. Here, a comparative and statistical text analysis of the linguistic construction of the kratēmata would shed further light on such theoretical considerations. Also, the hypothesis accepts a priori the synonymity of kratēmata with teretismos without taking into consideration the theory of the function of teretismos by Bryennios. This synonymity seems to arise from the use of the syllables te and re and the ancient verb τερετίζω (mimic the sound of cicada and birds) and the derivation of the word τερετισμός (teretismos) as the object of the verb. 38 This is a very common practice found generally in Greek traditional music and not only in ecclesiastical music, i.e., the χελιδονίσματα ('helidonismata', from the word χελιδόνι meaning swallow), a tradition originating from antiquity in ancient Greece. These were the songs of the New Year's eve of the spring, namely the first of March according to the Orthodox liturgical calendar at the time. These songs had clear references to the coming of Spring and the swallows. The custom is later found in Byzantium and transmitted orally under the year of the occupation by the Ottomans. The song was recorded in notation much later by Cl. Fauriel (1772Fauriel ( -1884 in his famous collection, "Greek Folk Songs". 39 Moreover, the origin of the ēchēmata has not been totally clarified apart from their appearance in the palace ceremonies with the use of the nonsense words, nana, hagia, and ananes, whose syllables, however, are not the dominant and most frequently appeared nonsense syllables in the kratēmata. The Byzantine scholar Christian Troelsgård has argued that the evolution of kalophōnia can be traced even earlier, to the early kalophonic tradition in 1300 AD, and this earlier melismatic tradition appears to have no less a melismatic character than the later. By analysing the kratēmata-like passages in the Grottaferrata manuscript (G), Troelsgård mentions that the kratēmata are regularly sung upon the nonsense syllables τερερε, τορορο, and in a few cases upon the nonsense syllables νε να as reminiscent of the intonation formulas. Αs an exception to this rule, in this old kalophonic composition, the vowels of the nonsense syllables are used in an even more expanded way than usual. Based on this observation, Troelsgård introduces the term meloform tropes for the kratēmata in order to support his hypothesis about the kratēmata as melodic expansions which gradually 37 Stathis   Stathis (2014) to the μεγάλες ἀργίες or μεγάλες ὑποστάσεις. 42 Second, as a specific melismatic composition compiled of nonsense syllables, it appeared for the first time in the codex EBE 2458 dated in 1336 (page 201ν: Κράτημα Κυρ Ξένου καί λαμπαδάριου τοῦ Κορώνη, ἦχος πλ. ά, Τοτοτο and page 203ν: Κρατήματα κατ› ἦχον, ἦχος ά, Ερερετερερε) 43 , and is then found in the majority of Byzantine liturgical manuscripts between 14 th and 19 th century AD. 44 Regarding the appearance of kratēma as neumatic sign, this is very well established by its use in thousands of known ecclesiastical music manuscripts, in Papadikai and other codices 45 , but regarding its function, Anastasiou mentions that the two types of kratēma (as sign and melic type belonging to kratēmata), do not correlate directly to each other. Rather, they are analogous because both constitute elements of prolongation, the former of a phonetic sign, the latter of the psaltic worship. 46 According to Wellesz, the sign kratēma "means a doubling of the rhythmical lengthening of the note under which it is placed, but it has a special cheironomic significance. It means a note produced with great emphasis." 47 Floros also talks about the "mega kratēma, a combination of diplē and petastē -the most important lengthening sign in Byzantine semiography". 48 Anastasiou indicates that the etymology of the word "kratēma" suggests the kratēmata as melismatic type intended exclusively for prolongation of the psaltic worship and melismatic ornamentation, a fact very often evident in indications in manuscripts for optional chanting of kratēmata. In fewer cases, the term kratēma is attributed to kalophonic composition because the kratēmata are structural elements almost of each and every kalophonic composition. 49 Anastasiou claims that the term kratēma is the predominant one but often is replaced -in order of frequency of occurrence in handwritten textsby equivalent terms with the same meaning as, ēhēma, teretismos, nenanismos, nai, prologos, logos, apologitari, apolytarisma, isophonia, katavasia, omonia (see  also Table 2). Anastasiou mentions that the term teretismos (or terirem) indicate kratēmata whose text is based upon the nonsense syllables terere, tototo, tititi, etc. Because almost all the kratēmata contain those syllables, the term teretismos can be considered as identical to kratēma. Anastasiou mentions that the word teretismos was not established in the ecclesiastical psaltic art but was already in use from antiquity, there meaning mimesis of the song of cicada or swallow, the song or the playing of the "kithara", or generally a type of trill. He continues by claiming that the term has a similar meaning (apart from this of the kratēma), of the mimesis of a bird, during 42 Anastasiou the Byzantine and post-Byzantine era. 50 Here, the scholar does not take into equal consideration the theoretical considerations of Bryennios about the musical function of the term, but he mainly keeps the linguistic part of the term teretismos (which coincides with the word's nonsense syllable te and re), in order to justify the argument about the identical terms kratēma and teretismos. His general claim, however, about the function of teretismos as mimesis of the voice of the cicada or the birds, offers some consideration of its role in the melismatic compositions, similar to Bryennios. For the latter consideration, he refers to sources from codices with texts and inscriptions mentioning that the swan "τερετίζει" or that the "τερετίσματα" are the song of cicada and swallow. 51 The notion of the secular origin of the kratēmata is not supported by him in the introduction of his thesis: Even if we accept the synchronicity of the psaltes (church singers of Byzantine music) with music instruments at ceremonies taking place out of the church, the completion of ecclesiastical compositions with instrumental melodies would presuppose the imperfection or incompleteness of the former. In a different case, such technical additions would be expelled from the ecclesiastical music when it would be found again to its physical place, the church. Or in any case, some melic compositions with their earlier form (without kratēma) and the later one (with kratēma), would have been delivered by the eponymous melodists. 52 Later in his dissertation, the author supports the impact of "θύραθεν" (secular) music on the kratēmata. According to this argument, the proofs of this affiliation are the national names of various kratēmata (e.g., Βουλγάρικον, Ἰσμαηλιτικόν, Περσικόν, Τατάρικον, Ροδαῖον, etc.), secular music terminology mainly in the post-Byzantine era (names of maqams in kratēmata, e.g., Segiah, Evitz, Atzem, etc.), and names of musical instruments for kratēmata which are considered "instrumental" or evoke the name of instruments (viola, nai, trumpet, etc.). Anastasiou rejects the fact that the "instrumental" kratēmata suppose the use of instruments or that the psaltes could mimic the sound and timbre of the instruments by performing the nonsense syllables of the kratēmata. Here, a degree of controversy concerning scholar's arguments relating to the "secular" origin of the kratēmata occurs as, initially, he rejects categorically this hypothesis and then "leaves the door open" for the impact of secular music on the kratēmata.
Anastasiou refers also to the opposite process, the influence of nonsense syllables of ecclesiastical music after the 14 th century AD on the nonsense syllables of the terenum (a vocal style which appeared in Turkey and North Africa), in addition to the use of Byzantine parasimantikē (notation) in post-Byzantine codices for the notation of secular music. 53 50 Anastasiou, Τα Κρατήματα στην Ψαλτική Τέχνη, 70-71. 51 Anastasiou,Τα Κρατήματα στην Ψαλτική Τέχνη,[70][71] Anastasiou, Τα Κρατήματα στην Ψαλτική Τέχνη, 79-¬80 (transl. by the present author from the Greek text). 53 Anastasiou,Τα Κρατήματα στην Ψαλτική Τέχνη,[445][446][447][448][449][450][451][452][453][454][455] The earliest Papadikē from 1336 AD contains kratēmata either as autonomous compositions or as parts of a composition. This date is set by Anastasiou as the terminus ante quem of the appearance of the kratēmata. 54 Concerning the origin of the kratēmata, Anastasiou argues that this is obscure and ambiguous. He mentions the spontaneous coming into existence of the kratēmata by pointing that they seem to appear suddenly in ecclesiastical manuscripts, both as autonomous melismatic compositions and as parts of these. 55 However, the explanation of automatic genesis seems not to be adequate. That's why, Anastasiou also proposes that the compelling appearance of the kratēmata should be a result of an evolutionary process, which we have to trace. Concerning this proposal about the origin of the kratēmata, the author does not formulate any new hypothesis but mainly agrees with the hypothesis by Stathis concerning the genesis of the kratēmata. Regarding their generative cause, the scholar claims that this was the inherent melismatic character of kalophonic melos, based on the use of intercalary consonants and the prolonging of the last melismatic syllable upon the use of nonsense syllables. 56 He states in the conclusion of his dissertation that the kratēmata were the result of the development of the melismatic character of the Kalophonic style and constitute an integral structural element of the kalophonic compositions. They appear at the end of the 13 th century AD originating from the nonsense and melismatic character of ēchēmata, sung by the domestikos as intonation formulas for each ēchos and for artistic ornamentation of the melos: They originate from the Kalophonic melos and the role of the domestikos to sing the ēhēmata or enēhēmata in the various melic compositions, in the beginning for the enforcement of the ēchos, and not much later for artistic ornamentation of those compositions. 57 Very soon, they are transformed into independent eponymous kalophonic compositions found in ecclesiastical Byzantine books such as Papadikai, Anthologies and Oikimataria, and later at the beginning of the 19 th century, in the Kalophoniko Eirmologio, an autonomous collection of kratēmata. Their gradual development and the accumulation of more and more kratēmata gave rise from the middle of the 16 th century to a new type of codex, the Kratēmatario. The melos of the kratēmata differs from the melos that appears in other types of the Papadiko genus in Byzantine music. 58 The kratēmata appear to be instrumental in character and melismatic freedom is allowed by the absence of the text. The performance and explanation of the Great Hypostases are simpler, shorter, repetitive, and individual and not in sequence and succession. The nature of kratēmata allows them to be influenced by the εξωτερικό (from outside) and mainly the οργανικό (instrumental / organic) μέλος, something that is reflected in the post-Byzantine kratēmata, the so-called ἐθνικά (national). During the transitional explanatory notation (about 1670-1814/15 AD) 59 , the exēgēsis (explanation) of the kratēmata is not attempted due to the fact that it does not offer practical significance since the chanting of the kratēmata is limited and almost abandoned during the post-Byzantine era, and the synoptic character of the θέσεις (positions) of the Great Hypostases of the kratēmata do not require specific interpretation. 60 59 Alexandrou, Παλαιογραφία Βυζαντινής Μουσικής, 310. 60 Anastasiou,Τα Κρατήματα στην Ψαλτική Τέχνη,[501][502][503][504][505][506] Hypothesis by Anastasiou Ambiguous origin of the kratēmata: Two hypotheses Figure 5. Historical hypothesis by Anastasiou (2005) Byzantium first half of 14th c.: sudden appearance of kratēmata • Specific melismatic composition compiled of nonsense syllables • Structural elements almost of each kalophonic composition • Melismatic interpolations and autonomous melismatic compositions • Kratēma is also neuma • Hints about echos of the secular music in the kratēmata • The term kratēma as a synonym for ēchēma, teretismos, nenanismos, etc., (see Table 2) Byzantium end of 13th c.: genesis of kratēmata by ēchēmata • Kratēmata as result of the development of the melismatic character of Kalophonic style • Integral structural element of kalophonic compositions. • They originate from the nonsene and melismatic character of ēchēmata

Conclusions
Touliatos constructs her hypothesis about the historical origin and function of the nonsense syllables in teretismata and kratēmata, mainly upon treatises from antiquity and Byzantium, as also upon sporadic other sources describing the historical role of those syllables. Touliatos's theoretical approach seems very solid in the parts dealing with the use of nonsense syllables in the music of ancient Greece and then the appearance of those syllables in the kalophonic style of the Byzantine, although the evolutionary process dealing with the emergence of the kratēmata by the teretismata, as also their distinctive musical features -apart from the ornamentation -is not explained sufficiently. Stathis bases his results almost exclusively on manuscripts of kalophonic compositions containing kratēmata and ēchēmata. He claims that the kratēmata originate from the ēchēmata and he uses the terms teretismata and teretismos as synonyms of the kratēmata. This hypothesis needs further confirmative investigation as it does not sufficiently explain the transition from the syllables of the ēchēmata to those of the kratēmata. Finally, the above hypothesis accepts a priori the synonymity of kratēmata with teretismos without taking into consideration the theory of the function of teretismos as described by Bryennios. Anastasiou distinguishes between the kratēma as neumatic sign in paleography and the kratēma as melismatic composition of the kalophonic style. In the second case, the kratēma, although it is the predominant term in use, is identical to the terms ēchēma, teretismos, nenanismos, nai, prologos, logos, apologitari, apolytarisma, isophonia, katavasia, omonia, and very often replaced by them. Concerning the origin of kratēmata, Anastasiou proposes their sudden appearance in the Byzantine chant or their genesis from the ēchēmata, although he suggests a deeper investigation of the hypothesis about 61 Touliatos, "Nonsense syllables," 239. 62 Stathis, The Anagrammatismoi and Mathēmata of Byzantine Chant, 111. 63 Anastasiou, Τα Κρατήματα στην Ψαλτική Τέχνη, 67-69.