Oxford Music Online
article url: http://www.oxfordmusiconline.com:80/subscriber/article/grove/music/44010pg4
Organ, §IV: The classical and medieval organ
IV. The classical and medieval organ
Since the 3rd century BCE it has been possible to regard the organ as an instrument composed of
four elements: (i) a wind-raising device operated by lever, pulley or other mechanism, directing air
under pressure to (ii) a ‘chest’ in which the wind is stored until admitted by (iii) a mechanism
operated by some kind of keyboard to (iv) one or more rows of pipes (see RANK). The absence of
any one of these elements prevents an instrument from being properly considered an organ. But
other instruments could well have presented models or given ideas to early organ makers,
particularly those in east Mediterranean countries. It is unlikely that at any single period the
hydraulic organ was so firmly established that builders were indifferent to the influence of such
wind instruments as the SYRINX, the MAGREPHA or the BAGPIPE.
The most comprehensive recent surveys of archaeological and documentary evidence relating to
classical and medieval organs are the studies by Jean Perrot (D(i)1965), K.-J. Sachs (C1970–80)
and Peter Williams (D(i)1993).
1. Greek and Roman antiquity.
No evidence, literary, iconographical, archaeological or even mythological, suggests that the pipe
organ existed before the Hellenistic period or originated in any other than the Hellenistic sphere of
influence. Later texts such as Athenaeus's The Sophists at Dinner (c200 CE) and Vitruvius's De a
architectura (1st century CE) accredit the invention of the HYDRAULIS to one man, Ctesibius, an
Alexandrian engineer and practical theoretician of the 3rd century BCE. Curt Sachs's assessment
(SachsH) of Ctesibius's achievement as uniting a mechanically raised and constant wind supply
to a set of panpipes is not a totally unreasonable conjecture; but the surviving accounts of his
work (written after his time) make it clear that he had also incorporated a wind-chest and even
some kind of keyboard. Thus the hydraulis has the essential features of an organ. That Ctesibius
was also said by Vitruvius to have invented a water clock offers an interesting parallel to the
makers of organs and clocks in the medieval cathedrals of western Europe: such makers were,
in effect, specialists in complex machinery. (See also WATER ORGAN.)
The principle of the water-pump is shown in HYDRAULIS, fig. But forge bellows were known much
earlier, and their power potential had already been described in the Iliad. Bellows could have
provided wind either directly to a regulator-chest under a row of pipes, or indirectly via the cistern
of a water organ. But there is no evidence that either of these was done before the 2nd century CE,
and it is possible that the organ was indeed born as a kind of engineering model, demonstrating
the efficiency of Ctesibius's wind-raising and wind-stabilizing equipment. Hero of Alexandria's
account (in On Pneumatics) gives no details of the pipes (whether flue or reed, open or stopped)
or what the material, size, compass, tuning, pitch or voicing were.
Vitruvius's musical interests are more obvious than Hero's. The ranks of his organ were made to
play separately by means of a specially constructed chest in which a channel ran lengthways
under each row of pipes, wind being admitted to the channel through a valve operated by an iron
Grove Music Online
Organ, §IV: The classical and medieval organ
14.3.2011 Organ, §IV: The classical and mediev…
…kb.nl/subscriber/article/…/44010pg4… 1/11
handle. The keys are returned to position by an iron spring. As the key was set immediately under
its pipes, either the close-set pipes or (more likely) the keys may have been unequally spaced.
The oldest reference to organ playing is a century and a half after Ctesibius: the ‘Delphic
inscription’ (90 BCE), full of implication about the organ's fame. Cicero, Lucretius, Petronius and
other authors also wrote of its powers. By the 2nd century CE the Roman organ was heard in
some of the more important theatres, games, amphitheatres, circuses, banquets and perhaps
processions; a 3rd-century Greek inscription at Rhodes even suggests that it was played in
Dionysian festivals. But the cylinder-pump water organ had so many disadvantages – requiring
precision engineering and good metal, yet difficult to maintain, move and keep from corrosion –
that it is easy to imagine bellows being applied over the years. Eventually, they replaced both
pump and cistern, but it is not known when, where and how. Even in the later Roman Empire,
however, organs were to be heard, and such poets as Claudian (c400) show organ playing to
have accompanied celebrations attending accessions to a consulate, weddings and banquets
during a period when ‘the singer has thrust out the philosopher’ (Ammianus Marcellinus, c350).
Inscriptions found in several provinces far from Rome (Arles, Colchester, Budapest, Asia Minor)
make it clear that organ playing was heard in gladiator contests.
The few 5th- and 6th-century references include one or two by early Church Fathers, particularly
those on the south and east coasts of the Mediterranean. But whether it was from personal
experience that such writers as Boethius wrote of hydraulic organs, or Cassiodorus of a bellowsorgan
with wooden keys, is not certain; nor has 20th-century research shown what music the
organ played, much less whether it played polyphony. Much can be tentatively conjectured from
the iconographical evidence (see HYDRAULIS, §§2 AND 3).
Parts of two Roman organs are said still to exist: fragments from Pompeii (now in the Museo
Nazionale, Naples) and major remnants of a small organ found in Aquincum, Hungary (now in the
Aquincum Museum, Budapest). But the Pompeii fragments, which seem to belong to two different
instruments, may not be parts of an organ, although their pipes are of cast metal, like those of an
organ. The Aquincum organ has a plaque dated 228 CE; its ‘reconstruction’ in 1959 somewhat
hampers more recent study of its parts. It has four rows of 13 cast bronze pipes, one open and
three stopped, and a wooden wind-chest, lined inside and out with bronze. Wooden keys operate
perforated sliders rather than pallets. E.L. Szonntagh's analysis (D(xvi)1981) appears to confirm
that, although the wind-raising mechanism is missing, the Aquincum organ was a hydraulis,
rather than a bellows-blown organ. Each set of pipes can be played separately from the others,
and Walcker-Mayer (D(i)1970) theorizes that the four ranks of pipes were tuned according to four
separate modes. Further research on this theory is necessary, but if the four registers of the
Aquincum organ did in fact produce modal scales based on different pitches, the knowledge will
have significant impact upon the understanding of the later development of modes, scales and
keyboards.
Reconstructions of the hydraulis suggest that it might have had a wind pressure anywhere
between 7·5 cm and 30 cm; also unknown is whether the pipework was always flue and, if so,
whether the diameters were constant. While written sources give no firm evidence, iconography
seems to suggest that: the pipes were usually flue; their diameter was constant; the tuning (in the
more complete examples) was probably multiple, providing a choice of modes for rather less
than an octave; and multi-rank chests may have provided different timbres with or without octave
and 5th-sounding ranks. But none of these conclusions is reliable. The Aquincum organ supports
the case for flue pipes, contradicts the suggestion that the diameter was constant, and leaves the
tuning and timbre uncertain.
2. The Byzantine organ.
By the end of the 5th century the new Roman Empire of the east, with its base at Constantinople,
had achieved a character of its own, intellectually conservative and favouring a world of abstract
thought far removed from the practical technology of ancient Alexandria and Rome. Although the
old Greek treatises were preserved in Byzantine copies and hence known to the Arabs,
engineering projects like organ making remained undeveloped for a millennium. But by the 8th
century western Europe itself no longer knew such masterpieces of Roman engineering as the
Vitruvian hydraulis. All the sources suggest that the European 9th- and 10th-century ‘organ revival’
14.3.2011 Organ, §IV: The classical and mediev…
…kb.nl/subscriber/article/…/44010pg4… 2/11
Hydraulis (centre) with horns,
cymbals, lyres, a psaltery and
lute:…
came about because the instrument was reintroduced from Byzantium.
Despite some hints in the sources, the organ was certainly not used in the Byzantine Church itself
(and indeed is still not except occasionally in churches in the USA). But at least two facts seem to
be clear: that most references relate to bellows-organs, and that the instrument continued to be
part of the secular, courtly pomp in the capital city. In the first connection, a 10th-century Arabic
source suggests that three (or two) bellows fed air into a large reservoir below the pipe-chest; in
the second, it was no doubt because of their use at banquets, chariot races, weddings,
processions and the like that organs were decked out in gold and costly decoration. Both the
‘blue’ and ‘green’ factions at court had an organ, but the instrument otherwise remained a rarity.
At his palace the emperor had both automata (the famous ‘golden tree’ with moving, whistling
birds activated by bellows) and true organs in which at least one emperor (Theophilus, 9th
century) took an interest. Nothing is known of the pipework, sound, compass, precise function or
repertory of the organ in the Great Reception Room, or indeed anywhere else, though one 9thcentury
source does refer to ‘60 copper pipes’ in what appears to have been a large table-organ.
Organs became objects of visual and aural show, eliciting wonder and respect as diplomatic gifts
or signs of royal power. In 757 a famous diplomatic instrument was sent to Pepin, King of the
Franks, at Compiègne. A monk of St Gallen (possibly Notker Balbulus) reported that the ‘King of
Constantinople’ also sent an organ to Charlemagne in 812, with bronze pipes, ‘bellows of bull
leather’ and three sound effects (rumbling thunder, trembling lyre, tinkling cymbali) possibly
suggesting pleno, flutes and little bells; but the source is doubtful, the language being somewhat
hyperbolic (or possibly psalmodic). In any case such instruments were not church organs but
extravagant gifts, like the 13th-century organ of 90 pipes sent from one Arab court to the Emperor
of China.
An event of evident importance in the 9th-century chronicles was the
arrival at Aachen in 826 of Georgius, a Venetian priest who
undertook to construct a hydraulis. According to a poem glorifying
Charlemagne's son Louis, Georgius's organ was a kind of royal or
national symbol of power: ‘The organ, not seen in France before, a
subject of pride for the Greeks, the only reason the people of
Constantinople felt themselves your master: even the organ is now
represented in Aachen’. Its intricate technology must have been the justification for such respect.
The Aachen organ was used for occasions of pomp, not for chapel services; the Utrecht Psalter
(compiled in France, perhaps near Reims) depicts, with little understanding, a hydraulis taking
part in an ensemble illustrating Psalm cxlix (fig.22); this too has little to do with church services,
being merely symbolic, or perhaps depicting some sort of ‘signalling’ organ.
3. The organ of the Arabs.
The high level of Arabic and Islamic culture from the 8th century to the 10th gave theorists and
craftsmen the opportunity to work on bellows-organs; theorists in particular knew of such
‘instruments’ but seem rarely, if ever, to have seen one. A famous source, the Epistle to (or from)
Muristus, describes two organs, one of which is a kind of siren or signal-organ; the sources
containing Muristus's writings are also interesting in that two of them (in Beirut and in the British
Library) show how a diagrammatic plan can become, under the scribe's hand, an unintelligible
pattern of abstract design.
Nothing is known of Muristus, and the graphic similarity of his name in Arabic to Qatasibiyus
(Ctesibius) was pointed out by Farmer (D(i)1931); Muristus appears to have been a Greek (or
Byzantine), and in any case derived his descriptions of instruments directly from Ctesibius's
Commentaries. But neither of them is a true organ. The first contains a chest of 12 pipes fed with
wind from the lungs of four men blowing through tubes into a regulator; the weight of the pipes
compresses the wind; the pipes themselves appear to be reeds, all of the same length but of
varying diameter and requiring different volumes of wind; the wind is admitted to each pipe
through a valve, presumably one worked by some kind of key. This seems to be the instrument of
‘formidable power’ referred to in the ‘Letter to Dardanus’ once attributed to St Jerome. The second
or ‘Great’ organ is a signal-organ perhaps not unlike the (smaller) magrepha and containing a
siren pipe or pipes blown at great pressure, used in battle by the Greeks, according to Muristus,
14.3.2011 Organ, §IV: The classical and mediev…
…kb.nl/subscriber/article/…/44010pg4… 3/11
Hydraulis (centre) with horns,
cymbals, lyres, a psaltery and
lute:…
or for similar purposes by other Middle Eastern peoples. The siren worked on the same principle
as the hydraulis, four pumps or cylindrical bellows providing wind pressurized by water in a
cistern.
There is no evidence that the organ became known again in western Europe through the cultural
activities of the Arab caliphate of Córdoba in Spain. But the possibility that this might have been
the case adds further importance to any work undertaken on this period in Iberian musical history,
for links may perhaps be discovered between Spanish–Arab instrument making and 9th-century
Benedictine musical life. In the eastern caliphates organs seem to have developed into mere
ingenious automata; but even in that state the Eastern organ seems not to have survived the fall
of Constantinople in 1453.
4. Early church organs.
The gift organs of the 8th and 9th centuries were Byzantine organs,
called ‘organum’ in the chronicles and still perhaps regarded as an
‘engineering contrivance’ rather than musical instruments. The
fanciful clock given to Charlemagne in 807 by an envoy of the
Persian king was Arabic; the Venetian priest Georgius, sent to
Aachen in 826, was possibly trained in Byzantium. Several Western
writers from the 8th century to the 13th knew of the Greco-Roman
organ but in most cases from vague written sources from which even the more astute authors got
the impression that the pipes were made to sound by water rather than by air – an idea sustained
by the Jesuit amateur physicists of a later century. The well-known picture in the Utrecht Psalter
(fig.22) of a hydraulic organ with two players and four alternating pump blowers is also based on
a misunderstanding of the Mediterranean hydraulis. Many of the early church writers refer to
organs in such hyperbolic or apparently unreal terms that their sources of ‘information’ must have
been in most cases literary. It is even possible that such references to organs with 100 pipes, like
that of St Aldhelm, were mistaken allusions to hydra, the 100-headed monster whose name is
the same as a documented abbreviation for the hydraulis. All ecclesiastical references to organs
before the 10th century are to be treated with caution, and even scepticism.
All these organs were secular. One of the great unsolved puzzles of music history is how and why
the organ came to be almost exclusively a church instrument in western Europe from about 900 to
about 1200. The early church was subject to two particular influences against any instrument in
church, and especially in the liturgy: the liturgy's origins in the Jewish synagogue, and Patristic
resistance to anything of profane or luxurious association. By the 9th century, however, the
intellectual and liturgical style of the church had changed. Like sung organum, the instrument
owed a great deal to Benedictine cultural centres, not only in their literacy and scholarship but
also in the opportunities which their large churches gave to the advancement of music. The
monastic revival in the late 10th century must itself have been a factor in the appearance of
organs, which had become ingenious objects for the use of the clergy, not the people. The organ
was never officially approved or even acknowledged in any known papal or pontifical document
despite the traditional legend that Pope Vitalian (657–72) introduced it. Nor, for one reason or
another, are any of the references to organs placed or used in church before the 9th century at all
reliable.
Organs, like tower bells later, were one of the irrelevances complained of by the new reformed
order of the Cistercians, judging by remarks made by St Aelred, abbot of Rievaulx in 1166; his
reference to the sound of the bellows, the tinkling of bells and the harmony of organ pipes is
highly reminiscent of older reports. St Aelred also referred to the crowd of people watching this
display as if in a theatre, ‘not a place of worship’, which suggests that organs were placed inside
buildings, perhaps for example a large Benedictine church. But all this does not necessarily
indicate that an organ was used during the service, nor even before or after the service. Perhaps it
was rather an object of curiosity, like a cathedral almanac clock. Other 12th-century sources imply
more clearly that an organ was used in some way during the services, perhaps for signalling
purposes, like bells at the Elevation.
Whether organs were used liturgically is not clear from the many 9th- and 10th-century references
to them. The notice describing the consecration in 972 of the Benedictine abbey of Bages, Spain,
14.3.2011 Organ, §IV: The classical and mediev…
…kb.nl/subscriber/article/…/44010pg4… 4/11
for example, makes it clear that an organ played near the entrance, ‘praising and blessing the
Lord’; but to surmise more is conjecture. Much the same could be said for the archiepiscopal
coronation at Cologne Cathedral in 950. Pope John VIII (872–82) wrote of an organ required ‘for
the purpose of teaching the science of music’, for which it remained useful to scribes writing
about and teaching musical proportions, for example at Benedictine centres such as Fleury and
St Gallen. The practical function of organs set up by, or in memory of, great abbots or landowners
is unknown; reference to organs used on feast days (e.g. in the Life of St Oswald, 925–92)
suggests if anything that they were extra-liturgical, a kind of church carnival object. The
Benedictine abbot Gerbert (Archbishop of Reims, 991–5) was said by William of Malmesbury to
have had a hydraulic organ put into the cathedral: an object of mechanical ingenuity, once again
coupled with a clock in the written account. Gerbert may have learnt the principles of the hydraulic
organ from the Arabs in Córdoba, where he lived for a time, since Benedictine manuscripts of the
period do not suggest any practical familiarity with the writings of Hero or Vitruvius. Nothing is
known of other 10th-century organs, such as that set up in Halberstadt Cathedral under its
Benedictine bishop Hildeward; nor are contemporary references such as those of Notker
Balbulus (d 912) helpful towards an understanding of the nature and purpose of organs. So many
of these writers were merely indulging in metaphor.
One detail of the Bages consecration of 972 was that the organ music ‘could be heard from afar’,
which may or may not imply that the organ was outside the church. But a large number of
references, second-hand or glossed though many must be, suggest that the organ was a loud
instrument by standards of the day. Is it possible to see the famous late 10th-century organ in
Winchester Cathedral as a signal-organ, used on feast days to summon the congregation or
overawe them (perhaps before or after services)? This does not preclude its having keys and
some musical potential; ‘signal-organ’ simply describes its loud tone. If the Winchester organ
was placed near the west or south door (stone screens were not known until the next century, and
at Winchester only the nave may have been capacious enough) its use could hardly have been
liturgical. Nor is it easy to see how an organ could have been liturgical in a much partitioned
church of the type known to the later Cistercians.
The Winchester organ was built by about 990, some decades after the Benedictines were fully
established there and later than modern commentators have said. Details of it appear in a
fanciful verse letter written shortly afterwards by the monk Wulfstan. Much quoted, much translated
and much misunderstood, the poem speaks of 26 bellows and 400 pipes in ten ranks, with the
40 notes arranged as two sets of 20 keys played by ‘two brethren of concordant spirit’ (see
BicknellH for an up-to-date interpretation). Each key was a perforated slider pushed in and
probably pulled out – hence the need for two or more players. Clearly some kind of organ did
exist; but there are good reasons for distrusting Wulfstan's account: despite the fanciful
references by St Aldhelm to what appear to be 100- and 1000-pipe organs, there are no other firm
details extant of such large organs, at Winchester or anywhere else; the numbers given for
bellows, blowers (70), pipes, ranks and keys are not plausible, whatever the diameter of the
pipes and however the wind was raised (even the number of players smacks of literary tradition or
at least of the poorly drawn hydraulis in the Utrecht Psalter); the general style and character of
Wulfstan's poem are those of an impressionable layman not concerned with technical accuracy
(for further details see McKinnon, D(xii)1974).
Three 11th-century sources are rather more practical than many later medieval manuscripts (see
§5 below). De diversis artibus, a large encyclopedia written in the first half of the century by the
German monk Theophilus, describes techniques used in making church objects – glass
blowing, painting, gilding, metal forging, bell casting and organ making. The sources of his
treatise leave its authenticity uncertain, the last part of it probably being a later compilation. A
second 11th-century treatise is the anonymous Berne Codex, a manuscript possibly originating at
the Benedictine abbey of Fleury; a third is a note by Aribo on pipe making.
Theophilus's organ (see fig. 1 [not available online]) could be mounted within a recess in an
interior wall, presumably at gallery level, with only its chest and pipes visible from the church and
these indeed covered by a cloth ‘tent’ when not in use. Later 13th-century screen organs would
have been equally well placed, in some cases better placed, when they came to serve as
alternatim instruments in the liturgy. Many details of Theophilus's organ are unclear, not least its
function in the church. Theophilus first advised his reader to obtain a treatise on pipe
measurement (lectio mensurae), which would presumably contain a table of concrete values or
actual pipe-scales, rather than mere Pythagorean ratios. The copper for the pipes was first to be
14.3.2011 Organ, §IV: The classical and mediev…
…kb.nl/subscriber/article/…/44010pg4… 5/11
Organ with sliders and a
conflatorium: drawing from the
Bible…
beaten very thin, then shaped around a gently conical mandrel, which suggests that foot and
resonator would be all of one piece, having the shape of a modern Trumpet resonator; pipes of
this shape are in fact depicted in some iconographic sources (figs.23 and 24 [not available
online]). The Berne Codex seems to suggest a more familiar type of pipe-foot, and gives ‘almost
4′’ as the longest pipe, but does not indicate the length of the foot unit. Theophilus's pipes are
equal in diameter, which may not be unreasonable in an organ of less than two octaves. In the
1980s Louis Huivenaar and Jan de Bruijn constructed conical pipes of the kind described by
Theophilus, placing them on a chest based on Zarlino's 1558 description of a chest from Grado,
Venice, believed to date from the 11th century, having 15 keys, two ranks of pipes, and spadeshaped
keys not unlike those in fig.24 [not available online]. This experiment suggests that the
conical pipes had a strong and complex overtone content, producing an extraordinary vocal quality
and seemingly well suited for playing a chant melody.
In his section on forging Theophilus described bellows, and from
other sources of the period it seems that such bellows were large,
capacious, and planned to compensate for leaks between feeder
and pipe. The feeders direct wind into a conflatorium or receiver,
shown in some 11th- or 12th-century miniatures such as the
Harding Bible (fig.23) and Cambridge Psalter. In the Berne Codex
the valve preventing the return of wind when the bellows are refilled
is placed in the collector, while Theophilus's valve is in the head of
each bellows. The main duct can be curved or straight (but perhaps
not mitred) and is usually shown as generously proportioned. The
keys of the Berne Codex organ closely resemble Hero's, consisting
of a ‘square’ depressed at one end, pushing in a perforated slider
(to which it is attached) at the other, and pulled back by a horn-spring to which it is tied. By the
13th century, according to a miniature in the Rutland Psalter (fig. 24 [not available online]),
organists were using their fingers separately (and rather elegantly) to depress the keys, which in
this miniature were broader and more substantial than some reproductions of it suggest.
5. Medieval organ theorists.
In the absence of any known organ remains between the 3rd-century organ of Aquincum and the
(?) late 14th-century positives of Sweden (see §6 below), historians must turn to the body of
‘medieval organ pipe theory’, readings of which have led to some misleading ideas about
medieval organs. The many sources have been seen as ‘treatises on organ building’ (Frotscher,
D(i), 1968; Mahrenholz, C1938; Fellerer, D(i)1929) or ‘treatises on pipe measurement’ (Perrot,
D(i)1965); but after 1966 researches into the now completely collated texts (see K.-J. Sachs,
C1970–80) have led to a new assessment of their purposes.
The texts, in some cases only a few sentences in clerical Latin, fall into three main categories.
The largest group (about 30 texts in 155 sources from the 10th century onwards) are those
concerned with the length of organ pipes calculated by ratios from an ‘initial’ pipe, itself of no
specified length; most of the length measurements take account of END CORRECTION which, in the
case of a row of pipes of the same width and mouth shape, is constant. A smaller group of texts
(11, in 11 sources) is concerned with the width or diameter of organ pipes, ignoring end
correction in calculating the length; some of these discuss the relationship of mouth width, cut-up
and foot-hole to the pipe diameter; none dates from before the 14th century. Neither of these two
groups covers the whole subject, since in fact variable pipe-widths and quasi-Pythagorean
demonstration of end correction are mutually exclusive. The third group of texts (three only, all
11th-century) deals with technical pipe making. These texts are Theophilus's De diversis artibus
(bk 3, pp.81ff), Cuprum purissimum (the Berne Codex), and the section ‘Sicut fistulae’ on pipe
making from Aribo's De musica. Some aspects of the organs described in this last group of
sources have received attention in §4 above.
The ‘pipe-length treatises’ rarely offer concrete usable measurements, nor do they outline any
pattern of values in which practical experience may have had a hand. Instead, the scalings
concern proportional values corresponding to the Pythagorean ratios known from monochord
theory. On the one hand, it is obviously possible to make an organ without determining the
acoustical phenomenon of pipes; on the other, no careful measuring of pipes leads to usable
14.3.2011 Organ, §IV: The classical and mediev…
…kb.nl/subscriber/article/…/44010pg4… 6/11
Organ with sliders and a
conflatorium: drawing from the
Bible…
pitches without proper tuning. Many treatises so resemble the numerous scaling texts for the
monochord and cymbala that the significance of their pipe-scalings should not be interpreted in
isolation; for pipes, strings and bells might have been cited primarily as examples of Pythagorean
ratios according to which a pipe approximately half as long as another will sound the octave
above, one approximately two-thirds as long the 5th above, and so on. Comprehensive instruction
treatises covering such matters include the works of Notker Labeo, Aribo, Engelbert of Admont,
Hieronymus de Moravia, Walter Odington and Giorgio Anselmi; an important branch in the
tradition was the widely known Scolica enchiriadis of the late 9th century. In no way were such
sources recipes for making instruments; rather, they outlined the kind of number theory which
theorists since Boethius had applied to music.
Both Theophilus and the writer of the Berne Codex were dependent
on ancient accounts, namely those of Vitruvius and Hero. Aribo's
account probably refers back to a manuscript tradition around the
uncertain figure of Wilhelm of Hirsau, who seems indeed to have
been concerned with actual pipe measurement. Most of the copies
of a text ascribed to him are provided with drawings showing the
scale of the first pipe (not unlike the measure line in Schlick's
Spiegel, 1511). But in other writers, end correction, the very factor
‘disturbing’ the neat theory of Pythagorean ratios, was itself
determined proportionally, calculated as a fraction of the diameter.
For such calculation the diameter was assumed to be constant;
hence the frequently repeated conclusion that the medieval organ
builder made a rank of pipes all to the same diameter. Optimistically interpreted iconography has
been seen to support this idea. But it should be remembered that the general medieval approach
to making things (i.e. before print technology brought craftsmen gradually to depend on visual
models) weighs against the practical significance of written-down treatises. Only two of the texts
cover organ building as such, and they are partly derived or even (in the case of Theophilus) the
result of a compilation. Moreover, practical details such as the remark in the Berne Codex that
pipes follow the modern diatonic genus (‘si … sit diatonicum genus quo maxime decurrent
moderne cantilene’) do not necessarily indicate an actual organ such as might be used in
liturgical music. The Sélestat manuscript (11th century) and the Berne Codex describe pipechests
of seven notes, and the former seems to make it clear that its three ranks are unison,
octave, unison; at the same time, an 11th- or 12th-century miniature, in the Harding Bible, shows
a keyboard of C, D, E, F, G, A, B , B♮, a set of keys showing one each of the known notes (see
fig.23). But it is not known whether these treatises and miniatures reflect more than certain
literate, second-hand and even non-empirical traditions passed on, perhaps indirectly, to their
scribal ‘authors’.
6. The church organ, 1100–1450.
Rarely in music history is conjecture taken more confidently as fact than in this area. Despite bold
and apparently plausible modern assertions that playing in 4ths and 5ths was known by 9thcentury
clerical organists, that alternatim chants were known in the Mass during the 11th century,
or that large organs played the puncti organici (and even the quicker upper parts or voces
organales) in the Île-de-France organa of the 13th century, there is no irrefutable evidence to
support them. It may be reasonable to assume that in the larger Benedictine abbeys (St Gallen,
Metz, Benevento) polyphony, organ playing and troping of plainchant were all linked; but it is not
known during which century the more cosmopolitan of the abbeys may have begun to use the
organ more integrally during Mass than they were ever to use their other expensive mechanical
equipment such as bells or clocks. Nor are technical matters concerning the structure of organs
any more certain. There is no evidence that, between the 10th and the 12th centuries, octave- and
5th-speaking ranks were used in abundance, or that reed and stopped pipes were also known,
as more than one modern writer has claimed. Much later still, basic assumptions are unreliable.
Iconography by no means establishes that organists had to use all of each hand to thump the
keys, at this or any point in organ history. Nor are archives less equivocal; church accounts do not
prove that the ‘little organs’ sometimes mentioned from about 1390 onwards were second
manuals of large organs or that, if so, such manuals were placed together or had the same pitch.
The use of multiple organs in large churches is well documented from this period onward, and
14.3.2011 Organ, §IV: The classical and mediev…
…kb.nl/subscriber/article/…/44010pg4… 7/11
the ‘little organs’ usually prove to be separate instruments from the ‘great organs’. Possibly the
second keyboard, up until at least the time of the Innsbruck Hofkirche organ (1550; the oldest
extant two-manual organ), should be seen rather as an extension to the compass of the first.
Organ research from about 1960 has been directed towards a circumspect interpretation of the
evidence, and a new period of re-examination concerning the evolution of the organ is inevitable.
Certainly the period 1100–1450 was one of great activity. During the 11th century more organs are
known to have been in monastic churches throughout western Europe; they were played at
ceremonies (probably outside the liturgy) and succumbed to the fires that frequently swept
medieval cathedrals (Canterbury 1114, Freising 1158, Merseburg 1199) – which suggests
perhaps that they were fixed in place. Some literary sources imply that the organ was played
during Mass, for instance the Roman de brut (c1155, Normandy):
"Quant li messe fu commensie …
Mout oissiés orgues sonner
Et clercs chanter et orguener"
– but such references are vague and merely image-evoking; poets’ sources were usually other
poets. More authentic sources of the 9th and 10th centuries suggest, however, that sequences as
well as the Te Deum were the most open to polyphonic vocal treatment, just as later they were the
movements most closely associated with the organ. A small portative and a psaltery are shown in
a 12th-century miniature but no ecclesiastical function is implied, any more than for the portatives
illustrating psalms in earlier psalters (e.g. the small organs hanging on willow trees at Psalm
cxxxvii in the Stuttgart Psalter, 10th century). But by the 13th century all instruments other than the
organ were excluded from various churches in Spain, Italy and France. The phrase ‘great organs’
is found in church documents (e.g. Erfurt Peterskirche, 1291), and by 1296 one French bishop
referred to the organ sounding five times in connection with the Sanctus – perhaps as a signal
rather than for music as such. There is no evidence that it played the tenor in Sanctus movements
or in any motet following at that point during Mass. But by the end of the 13th century secular
cathedrals from Exeter to Prague, Barcelona to Lübeck, were as likely to have organs as the
larger abbey churches. Whether erected on screens (as in England) or hanging on an upper wall
of nave or quire, the organs were usually located near the cantores, i.e. no longer near the west or
south entrances nor specifically near the main altar. It is not known, however, when large organs
were fixed in Theophilus's manner, and illustrations for psalm texts usually show much smaller
organs in ensemble. The phenomenon of the smaller fixed organ attached to, associated with,
and in some cases paid for by specially bequeathed chapels belongs to the 15th rather than the
13th century.
The large organ seems to have been an exclusively ecclesiastical instrument from the 9th-century
Western Church to 17th-century Italy. Probably by the late 13th century the cathedral or abbey
organ was occasionally used in alternatim music with the cantores, though presumably not with
the congregation itself. Jovannes de Florentia referred (c1350) to performance ‘partim organo
partim modulatis per concentum vocibus’. Early 15th-century keyboard repertory extant in the
Faenza Codex (I-FZc 117; see SOURCES OF KEYBOARD MUSIC TO 1660, §2 (i)) complements such
explicit references as the mid-15th-century Castilian rubric ‘the organs played one verse and the
clerics sang the other’. 14th-century documents usually suggest that whatever the organ played, it
did so on traditional church or local feast days, for example at Halberstadt Cathedral on
Christmas Day, for Easter Week, Sunday after Easter, Kreuzerfindung, Reliques of St Stephen,
Ascension, St Peter and St Paul, Dispersal of the Apostles, Mary Magdalene, St Stephen and St
Sixtus, Assumption, Patron, Nativity of Mary, St Michael, St Gall, All Saints and 12 other feast days
including Trinity and Annunciation. For three centuries organs were used only on feast days. But
by the end of the 13th century some churches had decreed against other instruments (Milan,
1287); by the 14th, alternatim performances took place, especially in the Office; by the early 15th,
many areas, such as the Upper Rhineland, north-central Germany, some English and Italian
cities, and the stretch from Rouen to Utrecht, had organs in most of their larger churches, and the
future of the instrument was completely assured.
It is impossible to trace this history step by step, despite a certain amount of archival, musical and
iconographical evidence. But certain general points can be made about the 14th and early 15th
centuries. Organs became known in cathedrals less as an exception and more as a norm; by
14.3.2011 Organ, §IV: The classical and mediev…
…kb.nl/subscriber/article/…/44010pg4… 8/11
Organ, c1380 (pedal pipes
1718), perhaps from
Abondance Abbey (Savoy),…
1425 the large positive (with front pipes arranged from left to right) was usually distinct from the
fixed church organ (with front pipes in mitre form with a set of larger pipes to each side, thus
requiring a rollerboard). All the evidence suggests that only open metal flue pipes were known,
though some commentators have seen such references as ‘plom … per las horguenas’ (church
of the Cordeliers, Avignon, 1372) as evidence that lead pipes were used for distinctive tonecolour;
lead was in fact the most common pipe metal in all countries during this period. Larger
organs in certain areas (Normandy and later the Netherlands, possibly England also)
occasionally had Trompes during the period 1390–1450 (i.e. a set of ten or so large open metal
Bourdon pipes, possibly played by a separate manual or pedal keyboard and placed to one side,
or both sides, of the main organ). Presumably they also had a BLOCKWERK, although apart from
the number of pipes in a few famous examples (e.g. 2000 at Amiens in 1429) little is known in
this regard before 1450. Presumably the pitch of their compass, whether from (apparent) B or any
other note, was roughly equivalent to men's voices, the total compass perhaps divided up and
distributed over more than one keyboard. However, so many significant unknowns are raised by
such summaries that describing the church organ before its more clearly defined types of 1450 is
mostly a matter of citing facts about individual instruments.
The organ at Sion, Switzerland, is usually dated about 1380
(although it may in fact be later) and has been much rebuilt. Despite
opinions expressed on its tone, and although some of the original
pipework seems to be incorporated in the present organ, nothing is
certain of its original sound, disposition, compass, pitch, voicing,
pressure, bellows, position, purpose or provenance. Nevertheless,
its case (fig.25) shows interesting elements: it has the typical shape
for such instruments, with the central mitre lines (like Arnaut de
Zwolle's organ at Salins); the castellated ‘towers’ to left and right
overhang the sides; and the wings (painted and perhaps made
about 1434–7) enclose the pipes completely. At Bartenstein in East
Prussia parts of an organ dated about 1395 existed before World
War II. The organ had a large chest for 27 keys (?FGA–a′) with three divisions for a large chorus of
nine (bass) to 21 (treble) ranks, case-pipes of 16′, and Principals 8′ + 4′. An ingenious
reconstruction of the chest was sketched by Karl Bormann (D(xv)1966) but little is certain,
particularly the stop mechanism whereby wind was admitted to chorus and principals at will;
perhaps the device was made not in 1395 but one or two centuries later. The organ at Norrlanda
(c1380), now pipeless and in the State Historical Museum, Stockholm, is a large positive with a
putative Blockwerk of three to six ranks. A set of 12 rollers conveys not only both pedal and
manual key-travel to the larger pipes held in small side towers but also the action of certain pairs
of keys (C /c , D /d , F /f , G /g ) to a single pallet. This is so sophisticated an arrangement,
not least in its resulting chromatic keyboard of nearly two octaves (C–a or c–a′; fig.26; for a closeup
view of the keyboard, see KEYBOARD, fig.), that doubts too must arise about the age of the
organ – which in any case appears to have case-work constructed out of panels from some older
choir stalls. An organ of surprisingly similar appearance is depicted in the mid-15th-century
stained glass of the Beauchamp Chapel, St Mary's, Warwick.
Extant 12th- and 13th-century church accounts merely record the presence of an organ; about
many areas of Europe, curiously little is known. Only during the 15th century were the great Gothic
churches of some areas constructed (e.g. in the Netherlands), but many were immediately
provided with an organ as part of the regular furniture. The first real details of church organs occur
in such documents as builders' contracts from about 1390 onwards, when for reference
purposes the anonymous scribe would distinguish the ‘opus maius’ from the ‘parvum opus
organum’ (Utrecht Cathedral, suggesting either two separate organs or an organ with a
Rückpositiv) or the ‘principaulx’ pipes from the Bourdons (Rouen Cathedral, suggesting Trompes
and other major Fourniture ranks), or even by 1420 ‘cinch tirants’, suggesting five separate stops
in a large positive (Aragonese royal chapel). Otherwise it was enough for an organ to be entrusted
to the craftsmen concerned, who had merely to see that it was ‘decent, good and to the honour’ of
the church (S Giovanni Evangelista, Venice, 1430).
Henri Arnaut de Zwolle, writing in the 1440s, described several organs he knew, including those
at Salins (c1400, Blockwerk of 6–15 ranks) and Dijon (c1350, 8–24 ranks); an account of his
treatise is given below (§V, 1). The most famous 14th-century organ is that of Halberstadt
Cathedral (c1361, rebuilt 1495), described in some detail by Praetorius (Syntagma musicum,
14.3.2011 Organ, §IV: The classical and mediev…
…kb.nl/subscriber/article/…/44010pg4… 9/11
Positive organ: engraving of an
organist and his wife by…
2/1619). The four keyboards were as follows: I, called Diskant by Praetorius, playing the plenum
(case pipes + Hintersatz Mixture; see FULL ORGAN), B–c′ (14 keys); II, also called Diskant, playing
case pipes (Prinzipal) only, same compass; III, called Bassklavier, B–a (12 keys, long protruding
levers perhaps worked by the knee, playing the 12 large bass pipes); IV, pedal keyboard, same
compass as III, used with (perhaps pulling down the keys of) the top manual. The largest rank of
pipes was at the equivalent of 32′ pitch, the total number about 1192, from 16 ranks at pedal B to
56 at top manual a′.
Praetorius by no means understood the historical nature of such old organs, nor is it clear from
his report what in the Halberstadt organ dated from 1361, what from 1495. But it is probable from
his account that the Blockwerk had multiple ranks of octaves and 5ths such that the manual
disposition was approximately as follows: [not available online]. From the details given, the pitch
level seems to have been a′ = c505. Praetorius also described the sound of this Blockwerk (see
BLOCKWERK). 20 bellows supplied the wind, all presumably needing to be operated for the
plenum, though his drawing shows only two operators (see Keyboard, fig.).
Praetorius gave other details about organs he described as old, and his suggestions could be
the starting-point for organ historians. For example, he guessed that semitones appeared in
keyboard compass from about 1200 and pedals from about 1220, that by 1450 only open pipes
were known, but that spring-chests had been built by about 1400 and separate stops by about
1250. The first date is late by two centuries if it is a question of B only, perhaps a reasonable
guess if intended to refer to the first ficta semitone (i.e. other than B ), but early by at least one
century if all five semitones were meant. The date for pedals must be about a century too early.
The date for open pipes is probably correct, and that for spring-chests could be correct but is
probably a little early. The date for separate stops seems early by at least two centuries if it refers
to a full-size church Orgelwerk. Other details given by Praetorius are more certain, for example
that some keys were as broad as about 60 mm, that some keyboards had a compass of B–f′ or
c–a′ (diatonic only) and the curious-seeming statement that some early pedals played only the
bass notes. Obvious though the last may appear, the large Bourdons or ‘teneurs’ (Notre Dame,
Rouen, 1382) may in fact often have been operated by a keyboard played by the hands or even by
the knees. The term ‘teneur’ is evocative, but what it signifies is uncertain; perhaps the keys
played the long notes of a vocal composition or an Intavolierung; perhaps ‘teneurs’ meant merely
large pipes as distinct from small (‘menus’ at Rouen, 1382, ‘Diskant’ in Praetorius). Certainly the
playing of a cantus firmus en taille on the pedals is a later speciality of the 16th century. But
whatever ‘teneurs’ was meant to imply, builders of the period knew well how to fashion pipes of
various sizes and scale, according to Praetorius.
At the end of the 14th century, then, a large organ within the area Rouen–Utrecht–Magdeburg–
Orvieto might be presumed to have had a Blockwerk of anything up to 80 or more ranks with open
cylindrical pipes of metal, played by a broad-keyed manual of 16 to 22 notes, possibly with a
further keyboard playing Trompes with or without their own chorus mixture, and exceptionally with
a second smaller organ in some way connected with the first. Smaller but independent organs
may have had, by custom, a longer compass, smaller keys, and a Blockwerk of fewer ranks. Not
enough is yet known for generalizations to be made about the organ of about 1390 outside the
region specified above.
7. The 15th-century positive and portative.
Although the POSITIVE and the PORTATIVE each form virtually
separate subjects, they offer a useful gloss on organ history at this
point because each demonstrates a striking uniformity unknown to
the larger fixed organ, and each demonstrates the limitations of
iconographical evidence. Portatives were small portable organs
blown by bellows (often single but sometimes a pair) operated by
one of the player's hands (usually the left), and played by his other
hand on a keyboard of up to two octaves, composed often of touchbuttons;
the instrument would have one rank of pipes arranged in
one or two (very rarely three) apparent rows. Such are the highly
detailed and prettily finished portatives depicted by such painters as
14.3.2011 Organ, §IV: The classical and mediev…
…kb.nl/subscriber/article/…/44010pg4… 10/11
Copyright © Oxford University Press 2007 — 2011.
Memling (three examples, that at Bruges being the clearest: for see [not available online]
PORTATIVE) and the Master of St Barthélemy (two examples). Positives were blown by a pair of
larger bellows operated by a second person (fig.27), and were played by both of the organist’s
hands on a more or less chromatic keyboard exceeding two octaves (usually beginning at B) and
composed of short finger-keys; two rows of pipes would form one complete rank, often with
Bourdon (or drone) pipes pitched in the bass, perhaps an octave below. Some portatives also
had (shorter) Bourdon pipes. In all known cases the pipes were open and of metal; the scaling is
progressive and the diameters diminish, at least in the better depictions; cut-ups often appear
low and the scale narrow; unless chords of more than two notes were played, the wind supplied
by the hand bellows must have been quite adequate, though presumably low in pressure.
That paintings always leave problems of interpretation may be demonstrated by one of the bestknown
of all organ paintings, the Van Eyck altarpiece at Ghent (1432). Despite the beauty and
apparent precision of the picture, the pertinent section of which is reproduced in fig.28 [not
available online], there are several puzzles. The front pipes, though painted well, are not placed
naturally; the tips of the feet rest right at the front of the chest top-board, while each pipe corpus,
whatever its diameter, passes behind the supporting brace, itself, however, of constant thickness.
The feet of the inner row of pipes are placed almost without depth of perspective, all exactly in the
middle of those of the first row – despite the latter's perspective. Unless the keyboard ran no
higher than appears (blocked by the player's hand and arm) the two rows of pipes must produce
only one rank; yet if the keyboard continued up symmetrically (as far to the right as the bass goes
to the left) the organ would have at least 35 keys, implying a unique pipework of two nonchromatic
ranks. The line made by the pipe-tops corresponds neither quite to a diatonic nor to a
chromatic tuning, and the pipes in the bass seem unnaturally narrow in scale. Apart from these
problems of depiction, the painting gives no information at all on certain points, such as the
purpose of the latch-key on the lower left; if it is for a Tremulant, one might expect other evidence
of the period for such stops; if it is a stop mechanism operating a valve to the rear chest, one
must assume that there are other pipes not seen but making up the second rank; if it is a key to
operate Bourdon or drone pipes, the pipes should be in evidence. Such questions can be
answered plausibly enough, but only by conjecture, for comparisons with other instruments are
too distant to be useful.
Barbara Owen, Peter Williams
14.3.2011 Organ, §IV: The classical and mediev…
…kb.nl/subscriber/article/…/44010pg4… 11/11