The Office of St. Olav

Eyolf Østrem

Paper held at the IMS conference in Leeds, 1999

St Olav was the most widely celebrated Nordic saint, with feasts of the highest ranks throughout
the Nordic countries, throughout the Middle Ages. But there weren’t many traces of his
sainthood while he was alive. He was a warrior and a pirate—which made him quite rich—who
happened to be propagating christianity, and who therefore was turned into a saint after his
death in 1030. His canonization (performed by his own bishop, Grimkell) was primarily
motivated by his struggle against paganism, but when he died, it was sword in hand, in an
attempt to reconquer Norway, which he had lost to king Knut the Great—one christian king
against an other.
The celebration of Olav as a saint and martyr began almost immediately after his death on
the battleground at Stiklestad in 1030. Various sources describe how he was canonized and
translated to Nidaros one year and five days after his death. At that time, his body had more or
less crawled out of the ground all by itself, if we are to believe Snorre Sturlasson.
St Olav in England. But after this, the Norwegian sources are practically silent about Olav
for a century and a half. We must instead turn to England for the earliest traces of a cult of St
Olav. From around the middle of the eleventh century, i.e. only twenty years after his death, we
have evidence of several St Olav’s churches and his name is mentioned in several liturgical
books—three litanies, three mass prayers in the Red Book of Darby, and finally the so called
Leofric collectar, which contains a complete office for St Olav.
Why this cult in England, and even in the parts of England where the Nordic/Norwegian
element in the population was the weakest? To make a long story short, all the earliest English
occurrences of a St Olav’s cult can be related to a bishop Grimkillus. He was one of the bishops
that Olav had brought with him to Norway, and he was the one who canonized Olav. In 1038 he
was appointed bishop in Selsey, Sussex, which is the closest neighbour to Exeter, where Leofric
was bishop, and by all probability he was also the one who put together the office in the Leofric
collectar. Thus the English cult should not be seen as a response to a demand from Nordic
citizens in England—it is rather, if a poetic touch be permitted, a foreign rose, planted under
favourable conditions, in a firmly established church organization, but which gradually
disappeared when its gardener died.
St Olav in Norway. If we turn to Norway again, although there is evidence of a celebration
of Olav, most clearly attested to by the pilgrimage, which seems to have been quite extensive
from an early date, there is no hard evidence of a proper office in use during the first 150 years
after Olav’s death. This should however not be surprising, for several reasons: That’s what the
commune sanctorum was there for in the first place. For a new saint, the most important thing was
to be normal, not special. Furthermore, the institutional foundation in Norway at the time was
hardly of a kind that would be able to support the dissemination of a proper office. No clearly
delineated bishoprics existed before Olav Kyrre (king of Norway 1066–93), and the
responsibility for the existing churches was shared by the relatively few bishops and priests, who
ambulated between churches rather than being stationary at one place. Most priests and bishops
in Norway during the first century were foreigners—englishmen and germans—who were taught
in their own liturgical traditions, who brought their own liturgical books with them to the
country and to the churches they visited. The chances for a new office to find its way into this
maze and establish itself throughout Norway, seem to be slim. This is not probable until a fixed
chapter church has been established, where the training of new priests can be accomplished.
This kind of fixity developed gradually during the twelfth century, and in 1153 Norway
became a church province of its own, with an arch-bishop in Nidaros, Olav’s city. From around
this date we see an upsurge in the use of Olav as a symbol: as eternal king, eternal bishop, feudal
lord of Norway, warrant for just legislation, and, above all: national patron saint. All this makes
the cult of St Olav all the more important, but this is a role and a stiuation that is new around
the middle of the twelfth century.
The “Eystein” offfice. At this stage not only were the possibilities better for a more
specialized liturgy, but a need might even have been felt for a proper office for St Olav. A new
office was compiled around the 1170s by Eystein Erlendsson, the second arch-bishop of
Nidaros. It was a full-scale secular office of nine lessons and a proper legend. It is compiled in a
style that was becoming rather old-fashioned at the time: the song texts are in prose and not
rhymed, which was quickly establishing itself as the standard procedure for office writing at the
time. This is the office that was used in the Nordic countries throughout the Middle Ages and
which is represented in the sources.
Sources. The source situation for mediaeval liturgy in the Nordic countries is meager. No
complete books containing the entire office have been preserved, only a handful of text
breviaries, including the printed breviaries from the beginning of the sixteenth century, which
only contain the texts, and two antiphonals, which only contain the melodies. For the large part,
we have to rely on the fragments of books that were used as wrappers for account books in the
royal administrations during the sixteenth and seventeenth centuries. The collection in the
Swedish State Arhcive contains ca. 20.000 such wrappers, taken from a total of ca. 6.000 books,
mostly liturgical. So far, around 70 of these contain parts of the St Olav’s office.
This is a material which gives little or no support to firm conclusions about variant groups,
families, diocesal variants etc. Instead it is a random sample of variants, that can give rise to
questions concerning the status and character of musical objects in the high middle ages—
questions of interpretation of these sources and of reconstruction, which one will have to deal
with when working with this material.
The sources generally agree on the reading of the melodies—but agree with what? Well,
with each other. Some sources are more accurate than others—but in relation to what? Well, to a
more or less floating “landscape” of variants. The general impression one gets from an overview
of all the sources, is that there is a certain range of variations, within which all the sources are
contained, that most variants occur at specific places in the melody, but also that the sources can
not easily be grouped into families on the basis of the variants. It seems more like there is a
stock of variants available, and that each compiler/copyist chooses from this stock.
Transmission. The answer has to do with the question of transmission, in its two main
forms in the West: oral and written. During the past couple of decades, beginning with the
writings of Helmut Hucke and Leo Treitler, it has become a firmly established and largely
accepted procedure to discuss chant in relation to theories of orality and writing. This debate
has mainly turned around the hows and whens of the origin of chant in the dark ages before
musical notation. Less attention has been paid to the subsequent transmission of this chant,
which has been left completely to the paradigm of written transmission, although the problems
of transmission in this period are just as intrigueing and challenging as in the early centuries.
Leo Treitler has discussed questions of oral and written transmission througout his career.
In an early version his answer looks like this (figure 1):

Treitler here distinguishes dichotomously between a purely oral and a purely written
paradigm of transmission. In the written transmission, the acts of generation and dissemination are
held strictly apart, and the composer’s intention is perpetuated by the score, throughout all
realizations of it, whether through performance or through copying. In the oral transmission, on
the other hand, there is no clear distinction between the generation and the dissemination; they
both work together in what Treitler calls the “actualization of a traditional structure”. Every new
actualization is in principle a valid representation of the basic structure—no version is more
correct than the other, there is, and can be, no Urtext of plainchant. What is to be reconstructed
is not so much an object as an act.
The model is appealing enough, but it is not unproblematic. Treitler uses the term
“actualization of a traditional structure” to avoid the connotations of improvisation, such as
erratic, unplanned, “on the spur of the moment” creativity. But later studies in this area have
shown that even this phrase is too vague: the evidence gained from the earliest written sources
of chant shows that the main corpus of chant was already fixed as stable melodies before the
invention of music writing. This means that there is a distinction between generation and
dissemination even in the oral paradigm; what is transmitted is not a “traditional structure” but
more or less individual melodies based on a traditional structure.
Writing and memory. A central element of this interplay is the parameter of preservation, in
its two main branches writing and memory. It is obvious that in a purely oral paradigm, memory is
the only possible means of preservation. This lead another scholar, Kenneth Levy, to the
conclusion that musical notation must have been developed hundred years earlier than generally
believed (ca 800), since the huge amount of different and individual melodies at that time
couldn’t possibly have been preserved as accurately as it was by memory alone. During the last
decade, more precisely since Mary Carruthers wrote The Book of Memory (1990), the role and
function of memory in medieval times have been reevaluated. Carruther’s book has quickly
become a most important corrective to any theory dealing in depth with the intellectual life of
the Middle Ages, both regarding the dual, oral–written character of medieval intellectual life, and
regarding the role played by memoria in this setting. Carruthers claims that the medieval
(intellectual) culture was an oral/written culture, where writing was used, but where memory was
still highly valued and practiced, and where writers like William Ockam and Thomas Aquino
could write entire books and quote long passages from memory.
This description of the medieval culture can easily be transferred to chant as well: as a
predominantly oral culture where writing entered the scene and gradually expanded its role, but
never took over completely. We know from different sources that throughout the Middle Ages
much of the church service was done from memory, or at least that it was supposed to be that
way. In all the nordic countries there are regulations that state that a new priest should serve in
the metropolitan church for 2–3 years after his ordination “to learn the ecclesiastical office and
the sacerdotal ministry”.
At the other end is the role of writing. It seems reasonable to assume for music writing (or
more precisely: chant notation) in the Middle Ages, once it has been firmly established, a role
lying between the two paradigms in Treitler’s model—more than the equal of performance, as
different but equivalent actualizations of the Traditional Structure, but less than its role as the
carrier of the Score in the Modern Paradigm.
The Ideal Object. This is one of the really fascinating aspects of high medieval chant: its
position in a system which is not any longer purely oral but not yet as purely written as it was to
become later. The Olav’s office—or any song for that matter—may exist as many different
objects, objects on different levels with different status and characteristics:

1. Ideal object—in a platonic sense, the idea of the song, regardless of the way it is
performed
2. Written object—
3. Memorized object—
4. Sung/performed object
5. Heard object

The Score in the Modern Paradigm is a conflation of 1 and 2, as an Ideal Object made concrete
in writing, and which gives rise to sung, heard or memorized realizations, but where these don’t
have any effect on the Ideal Object. But in an oral/written setting? In a predominantly oral
culture one will tend to stay within the smaller circle sung–heard–memorized. If such a thing as an
“ideal object” exists at all, it is inextricably connected to this sphere, to the degree that a written
version will also be a function of it, even in the cases where one actually sings from a score. In
other words: in an oral culture the sounding object will be fundamental both for the written and
the “ideal”. This gives the following figure:

What about the situation of a culture that is no longer only oral, such as in the case of the Olav’s
office? Does the presence of a written version change the premises for the ontology of a
melody? The melodies of the Olav’s Office would never have been transmitted only orally, given
that it was composed towards the end of the 12th century. Furthermore, it does not emerge
from the pre-notational darkness as a “realization of a Traditional Structure”, it has been
composed—put togehter—at a certain time, in writing, as specific melodies with a certain line of
events which has been decided once—and for all?
For every musical system there will be a balance between how much of the musical
“content” is stored in the music itself and how much lies in common depositories that are not
directly connected with the melody in question, as tacit knowledge about the musical system itself.
Generally one can say that in an oral system, much more of the musical content will be of the
latter kind. Each individual song relates strongly to the entire musical system.
A clear example is the role of formulaicism. A lot of songs are built around formulae that
are specific for the genre and the mode of the song, and which to a certain extent even adjusts
itself to the text. Most attention has been paid to certain groups of graduals and tracts where
big chunks of music recur unchanged in song after song. But the same system is at work also in
antiphons of the office. For each mode there is an array of formulas that can be analysed
according to function—typically one type of formula leads from the finalis up to the repercussio,
one type is a way of extending the quasi-recitation around the repercussio, and one type leads back
to the finalis again.
Formulaicism is one element among others in oral transmission. What they all have in
common is that they are practices that facilitate the reproduction of a melody, and that limit the
possibilities of variation for a given melody in a given genre and mode under given
metrical/textual/syntactical conditions. The invention of music notation gradually made these
techniques obsolete, at least to the extent that their function was to facilitate memorization. But
against the background of what has been said earlier, it seems reasonable to extend the paradigm
of orality to music of a much later date, long after the invention of music writing. It seems that
not only the outer characteristics, but also the principles lying behind them, are preserved in the
new context of transmission (i.e. writing), reenforced also by the fact that the old system of
transmission (i.e. memory) lived on, alongside the new. The consequence of this would be that
even music such as the Olav’s office in one sense could be conceived as actualizations of
“Traditional Structures”, not as Works of Art given once and for all.