1. Overview
2. The Pre-Byzantium Era
3. The Origins of Byzantine Music
4. Notation
5. Psalmody and Hymnody
6. Later Byzantine Era
7. Post-Byzantine Era
8. The Reforms of Chrysanthos
9. From the 19th Century to the Present
1. Overview
Byzantine music is the medieval sacred chant of all Christian churches
following the Eastern Orthodox rite. This tradition, principally encompassing
the Greek-speaking world, developed in Byzantium from the establishment
of its capital, Constantinople, in 330 until its conquest in 1453. It
is undeniably of composite origin, drawing on the artistic and technical
productions of the classical age and on Jewish music, and inspired by
the plainsong that evolved in the early Christian cities of Alexandria,
Antioch, and Ephesus. In common with other dialects in the East and
West, Byzantine music is purely vocal and exclusively monodic. Apart
from the acclamations (polychronia), the texts are solely designed
for the several Eastern liturgies and offices. The most ancient evidence
suggests that hymns and Psalms were originally syllabic or near-syllabic
in style, stemming, as they did, from pre-oktoēch congregational
recitatives. Later, with the development of monasticism, at first in
Palestine and then in Constantinople, and with the augmentation of rites
and ceremonies in new and magnificent edifices (such as Hagia Sophia),
trained choirs, each with its own leader (the protopsáltes for
the right choir; the lampadários for the left) and soloist (the
domestikos or kanonarch), assumed full musical responsibilities.
Consequently after ca. 850 there began a tendency to elaborate and to
ornament, and this produced a radically new melismatic and ultimately
kalophonic style. Back to Top of Page^
2.
The Pre-Byzantium Era
In the centuries
before Constantine, there are no musical manuscripts-all the musical
evidence is late; we have no music which is datable with the appearance
of the liturgical hymn texts. But if our later musical sources have
preserved for us even the essential features of the melodies with which
these liturgical texts were first associated, they will enable us to
form an idea, however partial, of what the earliest stratum of Christian
music must have been like. The insoluble problem of Early Christian
music is: how can one make deductions from the evidence in our earliest
surviving musical manuscripts? To what degree does the music they contain
reflect that of an earlier period? "Throughout the early Christian world,"
writes Oliver Strunk, "in impenetrable barrier of oral tradition lies
between all but the latest melodies and the earliest attempts to reduce
them to writing."(+) While it may be possible to date an early musical manuscript,
it is virtually impossible to say how old the melodies in it are. The
entire question may be seen not so much in terms of a faithful melodic
preservation but rather as the degree to which traces of an ancient
model may be gleaned from our earliest notated sources.
A marked feature of liturgical ceremony was the active part taken by
the people in its performance, particularly in the saying aloud or chanting
of hymns, responses, and psalms. The terms chorós, koinonía, and
ecclesía were used synonymously in the early Church. In Psalms
149 and 150, the Septuagint translated the Hebrew word machol
(dance or festival group) with the word chorós. As a result,
the early Church borrowed this word from classical antiquity as a designation
for the worshipping, singing congregation both in heaven and on earth.
Before long, however, a clericalizing tendency soon began to manifest
itself in linguistic usage, particularly after the Synod of Laodicea,
whose fifteenth Canon permitted only the canonical psáltai to
sing at the services. The word chorós came to refer to the special
priestly function in the liturgy-just as, architecturally speaking,
the choir became a reserved area near the sanctuary-and the chorós
eventually became the equivalent of the word kléros.
For the earliest period, however, authorities are fairly well agreed
that the background of the worship service is to be found in Jewish
ceremonies of that day, and a large degree of continuity between the
worship of the Jewish and Christian communities cannot be doubted. What
holds for primitive Christian worship in general is no less true for
the earliest Christian music in particular. A strong case can be made
to support the belief that the background for the earliest Christian
music is to be sought in the music of the Hellenistic Orient, and more
specifically in the musical theory and practice of Hellenized Judaism
of that day. The Old Testament had a conspicuous place in the thought
and worship of the New Testament Church. Old Testament quotations and
allusions, especially from the Book of Psalms, abound in the literature
of the New Testament, and a comparison of the oldest Jewish liturgical
poems with those of Eastern Christians points to a relationship between
Syriac and Hebrew poetry, thus establishing the possibility of Jewish
influence upon Christian liturgical poetry. We know that cantors of
Jewish origin were often appointed, even attracted to teach Christian
communities the cantillation of scriptural lessons and psalmody. In
this, the ancient manner of oral tradition did not fail to show its
inescapable vigour.
There were, however, other issues at stake. Throughout antiquity, Christian
literature wrestles with many questions: Was music in the liturgy to
be tolerated at all? If so, what kind of music? Was singing to be executed
by the parish? Then there was the matter of the singing of women which
appeared to be a point of utter vigilance. The Bible rapidly became
the Book of books for Christianity. Jewish domestic psalmody was bound
to become the model fundamental to Christian ecclesiastical chanting
in which ethnic forces shaped local modifications over a rather wide
range.
One major difficulty is involved in identifying that which was musically
performed-in ascertaining just what was performed in a more or less
"musical" manner. A reason for this difficulty lies in the fact that
worship is often described in only a summary fashion, and rather general
terms are used. There is, moreover, as is only to be expected, a lack
of any precise musical terminology in New Testament writings.
There are some popular misconceptions about early Christian praise which,
perhaps, ought to be clarified. Many believe that music played a dominant
role in Christian gatherings of Apostolic and post-Apostolic times.
But, in fact, the New Testament itself offers very little evidence of
this, and in the earliest Church ordos of the second and third centuries,
the part played by hymn singing conspicuously lacks mention. Saint Paul
certainly exhorts the Ephesians to admonish one another in psalms, hymns,
and spiritual songs (Eph. 5:19; cf. Col. 3:16)-but this does not refer to the context of communal worship.
In the second century, Saint Justin Martyr talks about a united "Amen"
at the ends of prayers, but not about music. Some modern writers assume
that the earliest Christian churches were based on Jewish synagogue
nuclei and consequently adopted Jewish practices. But a reading a rabbinical
sources of that time discloses a very minimal use of music in the services.
We soon learn that the synagogues rejected the cultic sacrificial rites
of the Temple and concentrated almost exclusively on Scripture and homilies.
Even the Book of Psalms, which one would expect to be the natural song
book of both Jews and Christians, played a less significant role than
is generally imagined.
Back to Top of Page^
3. The Origins of Byzantine Music
Byzantine liturgical music did not come about in a cultural vacuum.
It has its origins in the desert and in the city: in the primitive psalmody
of the early Egyptian and Palestinian desert communities that arose
in the 4th to 6th centuries, and in urban centres with their cathedral
liturgies full of music and ceremonial. It is this mixed musical tradition
that we have inherited today-a mixture of the desert and the city. In
both traditions-that of the desert and that of the city-the Old Testament
Book of Psalms (the Psalter) first regulated the musical flow of the
services. It was the manner in which this book was used that identified
whether a service followed the monastic or the secular urban pattern.
In the desert monasteries psalms were sung by a soloist who intoned
the verses slowly and in a loud voice. The monks were seated on the
ground or on small stools because they were weakened by fasts and other
austerities. They listened and meditated in their hearts on the words
which they heard. The monks gave little thought to precisely which
psalms were being used-they were little concerned, for example, with
choosing texts that made specific reference to the time of the day;
that is, psalms appropriate to the morning or ones appropriate to the
evening. Since the primary purpose of the monastic services was meditation,
the psalms were sung in a meditative way and in numerical order. The
desert monastic office as a whole was marked by its lack of ceremony.
But in the secular cathedrals the psalms were not rendered in numerical
order; rather, they consisted of appropriate psalms that were
selected for their specific reference to the hour of the day
or for their subject matter which suited the spirit of the occasion
for the service. The urban services also included meaningful ceremonies
such as the lighting of the lamps and the offering of incense. Moreover,
a great deal of emphasis was placed on active congregational participation.
The psalms were not sung by a soloist totally alone but in a responsorial
or antiphonal manner in which congregational groups sang a refrain after
the psalm verses. The idea was to have everyone involved in an effort
of common celebration: there was no place here for individual contemplation.
Thus, it is not until the fourth century, when Christianity and paganism
collide as a result of Constantine's mass conversions, and when imperial
ceremony entered liturgical solemnity in new and vast cathedrals, that
music rears its formidable voice. And even then it did so under very
special circumstances, and not without considerable monastic opposition.
The monks of the desert likened tunes to demonic theatre, to false praise
and to idle pleasure, satisfying the weak-minded and those of little
faith and determination. But this does not mean that the monks did not
chant. Their rejection was of worldly music, musical exhibitionism and
the singing of non-scriptural refrains and chants. It was, in fact,
the monastic population that later produced the first and finest hymnographers
and musicians-Romanos the Melodist, John Damascene, Andrew of Crete,
and Theodore the Studite. And it was the monastic population that also
produced the inventors of a sophisticated musical notation which enabled
scribes to preserve, in hand-written codices, the elegant musical practices
of the medieval East.
But the emergent heretical movements of the fourth and fifth centuries
exploited the charm of music and enticed many away from Orthodoxy with
newly-composed hymns. They were so successful that the Orthodox were
forced to retaliate by using the same weapon. At first, only hymns found
in Scripture itself were permitted: the Magnificat, the Song of Symeon,
the Psalms, the Old Testament canticles, etc., but later the Orthodox
wrote troparia and kontakia based directly on the metrical and musical
patterns of the heretics' hymns. These early compositions were specifically
designed as processional pieces, for use in the streets and squares,
not in churches, and they involved full congregational or crowd participation.
Thus from the fourth century onward, music became an indispensable element
of worship. It underscored that fundamental concept of koinonia
or communio which was so vital and so real in the early Church.
It was the task of all present to sing, to participate in song, to respond
with one heart and one voice to the celebrant. Note that music was never
understood as a private, personal, devotional exercise (though this
is not entirely excluded); its function was communal; it identified
the popular element of liturgical celebration. For this reason, any
music used in church which focuses attention onto a particular person
or group, which forces another group into becoming passive listeners
and observers, is alien to the age-old tradition of the Church and to
the accepted perception of liturgy as an act involving all the faithful.
This is not to say that there were no soloists-there were indeed, but
primarily it was their duty to lead and to cue responses from the assembled
body of the faithful, and not to extemporize or to innovate.
How was this accomplished? There were two kinds of singing in the early
Church: an ancient Responsorial form and a later Antiphonal form. The
former began with the soloist's singing of the response, usually a selected
verse from a psalm. This served to give the pitches to the choir (made
up of the entire congregation) which then repeated the response. The
soloist followed by singing the verses of the psalm in such a fashion
that the melody used for each verse or half-verse ended with the same
notes that began the response. Receiving their cues in this manner,
the members of the choir repeated the response after each verse. This
subtle method of achieving musical unity, peculiar to the Eastern service,
obviously had its origin in the practical concerns of the performance.
With the advent of trained choirs, however, the need for these cues
would undoubtedly have disappeared, and they were probably maintained
primarily for the sake of their contribution to the overall musical
structure. The Antiphonal procedure required that the congregation be
divided into two, each with its own leader and each with its own refrain:
this time the refrain did not need to be from the Psalter. In this form
the Small Doxology was always added to the psalm as a final verse.
Back to Top of Page^
4. Notation
There were no notes to record music until after the 9th century. St
Isidore of Seville in the 7th century lamented the fact that the sounds
of music vanished and there was no way of writing them down. Only towards
the end of the first millennium was it felt that the singers' fragile
memories were not adequately conserving the sacred melodies that something
was done to fix the plainchants in writing.
Byzantine chant manuscripts date from the 9th century, while lectionaries
of Biblical readings with ecphonetic notation begin about a century
earlier. Fully diastematic Byzantine notation, which can be readily
converted into the modern system, surfaces in the last quarter of the
12th century. Currently known as round or middle Byzantine notation,
it differs decisively from earlier forms (paleobyzantine notation) in
that it represents an explicit technique of writing, accounting even
for minor details of performance. When reading the earlier, simple notation,
the singer was expected to interpret or realize the stenography by applying
certain established rules (generally unknown now but absolutely familiar
to him) in order to provide an accurate and acceptable rendition of
the music. The change to greater precision came about initially in response
to an urgent need: to capture the vestiges of an old and dying melodic
tradition then losing its supremacy in the face of more progressive
and complex musical styles. But the actual process of substitution from
the implicit to the explicit system is not easily explained, since mixed
traditions characterize notational procedures used in the Byzantine
world, each new manuscript revealing a variance, an inconsistency, or
a deviation. Broadly speaking, scholars have discerned two principal
paleobyzantine notations, of common origin yet distinct and contemporaneous
in their development: Coislin and Chartres (the names are taken from
two exemplars, MS Coislin and a fragment of MS Lavra Γ. 67, which
was formerly at Chartres).[*] Their origins are believed to lie in the ancient grammatical
accents, and they are comparable to the Latin staffless neumes.
Specifically, Coislin is a notation that chiefly employs a limited number
of rudimentary diastematic neumes (oxeia, bareia, apostrophos, petastē,
and klasma) independently and in combination, with the addition
of a small number of simple auxiliaries and incidental signs. Chartres
notation, on the other hand, is mainly characterized by its use of elaborate
signs that stand for melodic groups. Around 1050 these two primitive
systems terminated their coexistence, the former superseding the latter
and continuing its development until ca. 1106. Toward the end of the
century it succumbed to the totally explicit round method. The new system
embodied a uniformity that is inherent in any written tradition, but,
more than this, it established a number of influential precedents both
in manuscript transmission and in musical theory. It suppressed the
instability of oral tradition, and it countered the inconsistencies
of diverse musical practices. Melodies written in round notation developed
an aura of sanctity and became models for subsequent generations of
composers. One immediate result of this was the appearance of new music
books for soloists (the Psaltikon), for choristers (the Asmatikon),
and for both (the Akolouthia). But much more was involved in
the substitution of notations than a mere evolution to greater clarity.
Other changes were taking place in liturgical ordos and in performance
practices, and the advent of the round system satisfied the demands
placed on music by a new class of professional musicians (the maistores),
who naturally favored an exact method of writing that could capture
the nuances and elaborations of their highly specialized art. Marked
developments in the liturgical tradition, which had reached a culminating
stage by the end of the 12th century, gave the scribes an additional
incentive to provide appropriate musical material in newly edited choir
books.
Following an independent development and surviving until the 14th century
in a relatively unchanged state is the notation that was devised to
accommodate Biblical lessons: ecphonetic or lectionary notation. It
comprises a small set of signs that occur as couples, one at the beginning
and one at the end of every phrase in the text, presumably requiring
the application of different kinds of cantillation formulas. Like the
Coislin and Chartres systems, ecphonetic notation was of value for the
singer, who used it only as a memory aid; but complete reconstruction
of the melody line is impossible today.
Byzantine chant notation in its fully developed and unambiguous form
represents a highly ingenious system of interrelationships among a handful
of symbols that enabled scribes to convey a great variety of rhythmic,
melodic, and dynamic nuances. Certain signs called somata (bodies)
refer to single steps up or down; others called pneumata (spirits)
denote leaps. Five of the former group also carry dynamic value, and
when combined with the pneumata, they lose their step value but
indicate the appropriate stress or nuance. For example, the oxeia
(acute)
marks an ascending second with emphasis (usually denoted
by >). When placed with the hypsēlē (high), the
ascending fifth
, the oxeia loses its intervallic value
but has its dynamic quality applied to the new note. Standing apart
from these is the ison (equal)
, which asks for a repetition of the
note sung before. Another group of signs refers to the rhythmic duration
(note lengthenings), and another (the hypostases) to ornaments.
At the beginning of the chant, a special signature (martyria)
indicates the mode and the starting pitch. Therefore, in order to sing
from a medieval Greek chant book, the trained cantor (psaltes)
would work his way through the piece by steps and leaps, applying the
necessary nuances and durations as required by the neumes. To avoid
confusion, scribes frequently drew the somata and pneumata
in black or brown ink and the hypostases in red.
The introduction of neume notation in the 9th century had both positive
and negative effects for plainchant. On the positive side, it meant
that an authoritative version of a plainchant melody could be transmitted,
without alteration or deterioration, to other singers in distant places
that were unfamiliar with the tradition. On the negative side, it meant
that plainchant melodies had in effect become fixed once and for all.
What do I mean by this?
During the first nine centuries of Christianity, the Byzantine musical
tradition of plainchant managed to keep alive a certain improvisatory
fervour that was also manifest in the spontaneity of prayers and rituals
in the early Christian liturgy. Now, with some strokes of a 9th-century
pen, the plainchant melodies were caught in a rigid stylisation. They
became as if embalmed and their stylistic profiles conformed to 9th-century
and eventually, later, tastes. The old chants that originated as "sung
prayers" were henceforth crystallised "art-objects." Yet once the neume
notation was available to Byzantine Church musicians, it was impossible
to ignore its capabilities. And soon the notation became a force for
artistic experiment, since it gave composers a way to try out new musical
ideas, letting them ponder their novelties and circulate them for others
to examine and compare.
Thus, with a supply of graphic devices both to enshrine the ancient
melodies and to record new compositions, the Byzantine musician embraces
the art of composing. To begin with, this art meant something
a little different from what it does today. It was not just a matter
of thinking up fresh and novel sound combinations and putting personal
inspiration on display. Certainly the sacred texts were given a musical
dress that was designed to enhance their expression. But this was accomplished
largely without injecting the human creative personality.
Most early Byzantine composers were content to practise their craft
anonymously in the service of the Church. Their names are unknown, and
in their musical techniques a similar impersonality prevails. The early
chants tend to be built out of little twists and turns of melody that
everyone had heard and used for generations. The word composing
actually means putting things together, and that was essentially what
the Byzantine composers did. They arranged, adjusted and stylised from
a fund of age-old melodic bits and phrases that were active in the communal
memory. Therefore, when a "new" melody was created, it was often not
entirely fresh and original. More frequently it was a refinement of
some existing strains. It is for this reason I said earlier that impersonality
prevails not only in anonymity but also in musical techniques. Back to Top of Page^
5. Psalmody and Hymnody
Unlike the acclamations and lectionary recitatives, Byzantine psalmody
and hymnody were systematically assigned to the eight ecclesiastical
modes that, from about the 8th century, provided the compositional framework
for Eastern and Western musical practices. Research has demonstrated
that, for all practical purposes, the októēchos, as the
system is called, was the same for Latins, Greeks, and Slavs in the
Middle Ages. Each mode is characterized by the deployment of a restricted
set of melodic formulas that is peculiar to the mode and that constitutes
the substance of the hymn. Although these formulas may be arranged in
many different combinations and variations, most of the phrases of any
given chant are nevertheless reducible to one or another of this small
number of melodic fragments.
Both psalmody and hymnody are represented by florid and syllabic settings
in the manuscript tradition. Byzantine syllabic psalm tones display
extremely archaic features such as the rigidly organized four-element
cadence that is mechanically applied to the last four syllables of the
verse, regardless of accent or quantity. The florid Psalm verses such
as those for communion, which first appear in 12th- and 13th-century
choir books, demonstrate a simple motivic uniformity that transcends
modal ordering and undoubtedly reflects a pre-oktoēch congregational
recitative.
All forms and styles of Byzantine chant, as exhibited in the early sources,
are strongly formulaic in design. Only in the final period of the chant's
development did new composers abandon this procedure in favor of the
highly ornate kalophonic style. The most celebrated of these
composers, and one entirely representative of the new school, was the
maistor St. John Koukouzeles (fl. ca. 1300), who organized the
new chants into large anthologies. This final phase of Byzantine musical
activity provided the main thrust that was to survive throughout the
Ottoman period and that continues to dominate the current tradition.
Back to Top of Page^
6. Later Byzantine Era
Turning now to the later Byzantine period itself and on to our own times,
we enter the era in which music is something taken entirely for granted
in Christian worship: a feature automatically expected. To celebrate
a service without music would seem highly irregular. In a large measure
it is the event which many most look forward to because music has come
to identify the festive nature of a liturgical occasion-the aural embodiment
of that which has brought the faithful together.
How is it that music has taken over in this way? Why has it become the
measure of liturgical prayer and worship? It is precisely because it
is an art of great subtlety and power which, when used incorrectly, can
greatly distort or even caricature sacred poetry, but when understood
properly, it can heighten the significance of the celebration, contribute
to prayer, and emphasize the corporate nature of worship.
Music functions as a dramatic element-it has a unique and central place
in the general structure of liturgy; it has acquired liturgical significance.
Almost every word pronounced in church is "sung" in one form or another.
And the manner in which it is sung greatly affects the nature of the
service. Week by week, season by season, the Church's song draws out
the inner meaning of liturgical poetry.
Back to Top of Page^
7. Post-Byzantine Era
The year 1453 has been considered terminal by most writers, and while
none would flatly deny that traditional musical elements, both practical
and theoretical, were preserved at least until the middle of the sixteenth
century, most would uphold the view that the hymnodic productions of
the Ottoman era represent a disintegration of the authentic, Byzantine
forms of artistic expression and were the results of a growth of new
and innovative impulses that were alien to the spirit and evolutionary
pattern of the medieval past. As we look closer into the history of
Christian art in Ottoman times, we may detect in the literature a curious
duality: a mixture of conservatism and elasticity, of traditional compositional
methods and personal self-aggrandizement, of laconic control and specious
exoticisms. This duality is particularly apparent in the musical repertory
where both old and new are seen to exist side by side. A policy of artistic
liberalism and reverence for the past was the hallmark of the epoch.
For while resemblances to past practices stand out as both familiar
and apparent, it is also the differences manifested within the familiar
procedures that grant the absorbing attention and appeal experienced
in the music, and this becomes increasingly obvious the more we discover
the historical and technical processes and the origins and transmissions
of the compositions. Ultimately, each chant is unique is some particular
way and even a passing familiarity with the musical conventions of the
time, makes it possible for us to appreciate many of the individual
features. Collectively, these elements create a new musical vocabulary,
one which characterizes and eventually epitomizes an emerging neo-Hellenic
style. From an accumulated experience of these individual traits, our
knowledge of this style is more certain and we can begin to move with
more assurance to its proper interpretation and evaluation. Otherwise,
we shall forever be unable to fathom fully the sophisticated craft that
those diligent scribes from Constantinople, Mount Athos, Cyprus, Crete,
Serbia and Moldavia enshrined in collections which until today have
been undeservedly ignored.
One highly controversial
figure was the Cretan poet, theologian, calligrapher, singer, diplomat,
scribe and priest Ioannes Plousiadenos (born around 1429) who later
became Joseph, Bishop of Methone. After 1454, he was one of twelve Byzantine
priests who officially supported the union of the Eastern and Western
Churches ratified by the Ferrara-Florence Council of 1438 and 1439.
He even wrote the texts for two parahymnographical kanons, one entitled
"Kanon to Saint Thomas Aquinas"), which glorifies the great Catholic
theologian, and the other, "Kanon for the Eighth Ecumenical Council
which assembled in Florence"). The latter is modelled on the metrical
and rhythmical patterns of one of the Resurrection kanons in mode IV
plagal by Saint John of Damascus, but it was hardly likely to have been
used in the Greek Church because of its pro-henotic sentiments, triumphantly
celebrating the outcome of the Council of Florence at which Orthodox
acceptance of the "filioque" phrase in the Nicene Creed was allegedly
secured.
Very recently, evidence
has been discovered of Plousiadenos's involvement in musical composition
to serve the same end. In an attempt to introduce Western polyphony
into the Greek Church, Plousiadenos wrote at least one, or possibly
two, communion verses (koinonika) in a primitive kind of two-voice
discant. Apart from these isolated examples, the experiment with Latin
polyphony in the East had run its course, and inevitably so. It was
not until several decades later that the choral ison or drone-singing
was introduced into Greek church music, marking a fundamental change
from the centuries-old monophonic tradition. The earliest notification
of the custom appears to have been made in 1584 by the German traveller,
Martin Crusius.
A strong case can surely be made to classify the period of musical composition
from around 1500 to 1820 (the year when musical print replaced the handwritten
codex) neither as "post-Byzantine" nor "neo-Byzantine," nor even as
"Byzantine," but rather as neo-Hellenic, since the musical aspect of
artistic creation, particularly after the seventeenth century, participated
with other art forms in establishing a widely-acknowledged modern Greek
renaissance. Understood in this manner, it is less likely that one will
view the artistic and technical productions of the Ottoman years merely
as an extension of Byzantium or as its decadent and aesthetically inadequate
offspring.
At the forefront of this renaissance is sacred chant, the recorded history
of which is preserved in an imposing bulk of musical manuscripts (most
of them dated) that are located in widely dispersed and often inaccessible
collections: public, private and monastic. Despite the fact that it
may take a great many years to acquire a thorough familiarity with all
of the sources that are known today, it is yet possible for us to divide
the history of the evolution of church music from the fall of Constantinople
until the Greek revolution into five periods:
(a) 1453-1580 - a time
of renewed interest in traditional forms, the growth of important scribal
workshops beyond the capital, and a new interest in theoretical discussions;
(b) 1580-1650 - a period
of innovation and experimentation, the influence of foreign musical
traditions, the emergence of the kalophonic (or embellished)
chants as a dominant genre, and the conception of sacred chants as independently
composed art-objects;
(c) 1650-1720 - when
extensive musical training was available in many centres and when elegantly
written music books appear as artistic monuments in their own right.
Musicians of this age were subjecting older chants to highly sophisticated
embellishments and their performance demanded virtuosic skills on the
part of the singers. In addition, the first attempts at simplifying
the increasingly complex neumatic notation were being made;
(d) 1720-1770 - a period
of further experimentation in notational forms, a renewed interest in
older, Byzantine hymn settings, the systematic production of music manuscripts
and of voluminous Anthologies that incorporated several centuries of
musical settings;
(e) 1770-1820 - a time
of great flowering in church music composition and the supremacy of
Constantinople as a centre where professional musicians controlled initiatives
in the spheres of composition, theory and performance. Among these initiatives
were: further notational reforms, new genres of chant, the reordering
of the old music books, the more prominent intrusion of external or
foreign musical elements, and, finally, by 1820, the termination of
the hand-copied manuscript tradition. Back to Top of Page^
8. The Reforms of Chrysanthos
The decade 1810-1820 was, for the history of Greek chant, both turbulent
and decisive. Two major goals were finally achieved: first, the implementation
and universal acceptance of an entirely new notational system (1814)
which had evolved from the interpretative experiments of Balasios the
priest (flourished around 1670 to 1700) through the formulations of
the protopsaltes, Ioannes Trapezoundios (1756), of Petros Peloponnesios
(ca. 1730-1777), of Petros Byzantios (d. 1808) and of Georgios of Crete
(d. 1816); and second, as a consequence to the former, the invention
of musical print and the simultaneous publication of the first music
book (1820).
Chrysanthos of Madytos (ca. 1770 - ca. 1840), an uncommonly well-educated
and highly cultured hierarch, was primarily responsible for the reform,
and his system survives until this day. He had an excellent knowledge
of Latin and French, and was familiar with European as well as with
Arabic music, being proficient in playing the western flute and the
eastern "nay." Chrysanthos had learned the art of chanting from Petros
Byzantios and himself taught singing. As a composer and educator, he
became acutely aware of the need for more clarity in the process of
studying and understanding of Greek church music. The medieval neumatic
notation had now become so complex and technical that only highly skilled
chanters were able to interpret the symbols accurately. To facilitate
that end and to simplify the teaching of this difficult art, he invented
a set of monosyllabic sounds for the musical scale based on the European
sol-fa system but using the first seven letters of the Greek alphabet.
Each degree corresponded to one note in the scale:
Πα-Βου-Γα-Δι-Κε-Ζω-Νη = Rε-Mι-Fα-Sοl-Lα-Sι-Dο
In addition, he systematized
the ordering of the eight modes into three species: diatonic, chromatic
and enharmonic. Within each of these three categories, the intervallic
progression of the degrees was fixed according to elaborate mathematical
calculations. Chrysanthos also introduced new processes of modulation
and chromatic alteration and abolished some of the notational symbols.
As a result of these efforts, a large repertory of hymnody was made
available to chanters who were ignorant of the melodic and dynamic content
of the old signs.
Owing to this breach
with the traditional methods of teaching, Chrysanthos is said to have
been exiled to Madytos by order of the Constantinopolitan patriarch.
Yet, apparently this did not stop him from pursuing his highly original
approach to the teaching of ecclesiastical music. In Madytos, he found
that his pupils were able to learn in ten months what had formerly taken
ten years. The crucial device speeding up the process of learning appears
to have been his use of the aforementioned newly invented solmization
syllables. Finally exonerated by the Holy Synod, Chrysanthos was then
given a free hand to teach music as he saw fit. It was at this point
that he joined forces with the protopsaltes, Grigorios and the archivist,
Chourmouzios, both of whom seem to have had less formal education than
Chrysanthos, yet according to their biographies possessed a great natural
ability for music. All three taught at the Third Patriarchal School
of Music (opened 1815) and this ensured the success and propagation
of the new system. The results of Chrysanthos's research and teaching
methods appeared for the first time in a treatise entitled "Introduction
to the theory and practice of ecclesiastical music written for the use
of those studying according to the new method" published in Paris in
1821. Eleven years later there appeared in Trieste the more exhaustive
and highly influential Great Theory of Music which, in its first
part, expounded the new theories and notational principles of the three
reformers.
The second part of the Great Theory is purely historical. Chrysanthos
made an ambitious but unsuccessful attempt to present, in the form of
a chronicle, a general history of music from the time before the Great
Flood to his own day. It is recorded that he wrote many other works,
including transcriptions of Greek church music to European staff notation
and European music to the notation of the new method, but none survives.
Despite its numerous shortcomings, the oeuvre of Chrysanthos is a landmark
in the history of Greek church music since it introduced the system
upon which are based the present-day chants of the Greek Orthodox Church.
The invention of musical type marked the end of the long and fascinating
tradition of the music manuscript. In 1820, Peter Ephesios, a student
of the three teachers, published in Bucarest the editions of the Anastasimatarion
and Syntomon Doxastarion by Petros Peloponnesios. And, of the older
pieces, those that entered the printed repertory were randomly selected
by subsequent editors. After 1830, the official musical tradition of
the Greek Orthodox Church was represented by the following books: the
Anastasimatarion, the Heirmologion and the Syntomon Doxastarion of Petros
Peloponnesios, the Syntomon Heirmologion of Petros Byzantios, the Doxastarion
of Iakovos the protopsaltes, and the New Anthology of the Papadike-all
re-written according to the interpretations of Grigorios protopsaltes
and Chourmouzios in the new, simplified notation of Chrysanthos. Back to Top of Page^
9. From the 19th Century to the Present
The emergence of the printed music book after 1820 led to a standardization
of the chant repertory both on mainland Greece and on Athos. Selected
popular works of the great Constantinopolitan masters of the 18th and
early 19th centuries were type set and included in anthologies of chant.
But alongside these, simplified Western-style melodies were also making
inroads in popular editions of sacred music published, for example,
by the influential Zoe movement.
For a short time Athos could not resist the increasingly fashionable
Italianate style that was being introduced by Western trained musicians
and by the great influx of Russian monks on the Mountain before 1917.
But this was soon to be counterbalanced by the new sounds of the Asia
Minor refugees who flooded into Greece and eventually onto Athos after
the 1920s and 1930s-precisely when the Russian population on the Mountain
was entering a decline.
To begin with, the Church music of these Anatolians, though very much
a continuation of the earlier tradition of Ottoman times, was rejected
by the Greek urban middle classes as vulgar and "Turkish." They had
become enamoured of the sweet polyphonic choirs, some of them with organ
accompaniment. But, in time, radio, the gramophone and television also
proliferated sophisticated European styles-and these styles, though
in a neo-Byzantine dress, have affected certain repertories of Athonite
music even to this day.
Even as early as the 18th century there is evidence of a sharp
negative reaction by the Athonites to city church music. An anonymous
hand writes in a Vatopedi manuscript the following stinging remarks
in verse:
The
psalmodies of Byzantium
like the nightingales are heard;
While those of the Holy Mountain
resemble the tunes of guileless swallows;
But the ones in Athens
warble like the falcons;
And the psalmodies of Crete
are the arid squawking of the crows.
There has indeed been
a revival of traditional Eastern-style chant on the Holy Mountain, just
as there has been a revival of traditional icon painting. But wittingly
or unwittingly elements of Western diatonic music have blended with
the chant-a phenomenon reminiscent of what we had observed in earlier
centuries with the infiltration of Ottoman sounds into Byzantine melody.
Another feature of Athonite musical life in the post-war years has been
what I term the cult of the virtuoso. Until its very recent return,
choral music fell into a decline on the peninsula and instead one heard
master soloists improvising and elaborating chant with extraordinary
vocal skills and deft Oriental turns. The most famous of these soloists
was the deacon Dionysios Firfiris (d. 1991), whose evocative voice and
improvisational skills created a sensation both on and off the Mountain.
Since the mid-1970s, with the revival of monastic life by young, educated
monks, the musical emphasis has begun to shift from performance by an
individual to that by the group. For many years Simonopetra alone has
employed full double choirs for every service, each day of the year.
Its example has recently been followed by Vatopedi. This more traditional
performance practice is gaining popularity in convents and monasteries
on the mainland and abroad. Moreover, use of the Book of Psalms-the
ancient song book of the early monasteries-has been revived, and new
melodious settings for them have been composed.
Approximately fifteen years ago, a suave, lyrical melody set to a religious
poem by St. Nektarios of Aegina was composed by a monk at Simonopetra
and subsequently recorded on cassette tape and CD. Within two years
this melody circled the globe. It has captured the hearts of Orthodox
choir masters worldwide. The hymn, entitled, "O Pure Virgin," can today
be heard sung in Japanese, French, Tinglit, Italian, Russian, Swahili,
Arabic, Romanian, English, and many other languages. Its popularity
is entirely due to the fact that it combines familiar elements of two
different musical cultures: the harmonic and metrical features of European
lyrical ballads with the vocal production and exoticism that evokes
a flavour of the East.
What of the future? I believe that we shall observe a greater degree
of choral singing as opposed to soloistic virtuosity-though the latter
will not disappear entirely for some time. Athonite music will also
be greatly commercialised in the near future with the proliferation
of CDs and chant anthologies in countries beyond Greece. Such tendencies
have are already visible in Romania, Bulgaria, Serbia, the Middle East
and the United States of America. On the other hand, there has also
been a recent tendency to examine the old manuscripts in order to re-discover
earlier traditions and vocal practices. Western musical tendencies,
though perhaps never acknowledged as such, may continue to blend with
the chant.
The Athonite musical tradition has adapted over the centuries to changing
cultural tastes and conditions. This identifies it as an art that is
living and flexible. At all events, because of its prestige, Athos will
be a pace-setter for trends well beyond its own territory. Back to Top of Page^
(+)Strunk, Oliver, Essays on Music in the Byzantine World
(New York, 1977), p.61. (to reference)
[*] The background "wallpaper" of these webpages is taken from an
eleventh-century Sinai manuscript written in Chartres notation. (to reference)