Abstract
The precise nature of the aulos mouthpiece, especially in the Classical period and before, has been disputed: above all, was it a single or a double reed? A definitive answer to this question has the greatest consequences for the nature of aulos music regarding not only timbre but more importantly control of dynamics, phrasing, and microtonal adjustments. From a survey of iconographical, literary and archaeological evidence, it is argued that the instruments of higher cultural esteem were universally equipped with double reeds. Single reeds were known as well, but are almost completely eclipsed in our sources.
1 Introduction
An increasing interest in ancient Greek music has resurrected an old dispute over some puzzles that had already baffled the first researchers who came across finds of ancient Greek instruments well over a century ago.1 One of the key issues, which still stirs much controversy, concerns the type of the aulos mouthpiece. Although the debate on its nature and properties has been continuously going on, with varying dynamics, the arguments that have accumulated over time have never been collected in a comprehensive study.2 Recently, a revival movement of ancient Greek instruments has rather silently accepted certain arguments and accordingly adopted a specific kind of mouthpiece for playable reconstructions of ancient doublepipes. However, this has not put an end to the debate – on the contrary, the decision in favour of a certain kind of reed and its consequences for the musical character of the instrument is still being questioned by a dwindling number of supporters of different options. Since the arguments have sufficiently matured, it seems that the time is finally right to assemble and analyse available evidence.
Without doubt the aulos was the dominant wind instrument of Greek and Roman antiquity, filling public and private, sacred and secular spaces with its mesmerising timbre, accompanying solemn events – most importantly, rites of passage and offerings – no less than boisterous processions and jovial banquets. As a ubiquitous attribute of musical culture, the instrument acquired a profound symbolic meaning that was linked to its role as a vehicle for reinforcement of cultural identity.3 This important role was reflected in the large volume of visual and textual depictions, which had long formed almost the only surviving testimonies of the instrument’s bygone prestige. However, during the last two centuries finds of smaller and larger fragments as well as, sporadically, entire pipes have been recovered in archaeological excavations, which were paramount in determining the musical character of the instrument.4 Based on these finds, scholars were able to conclude that the aulos was an aerophone consisting of two cylindrically bored pipes played simultaneously, each of the pipes being sounded by its own vibrating reed mouthpiece. What has been much more difficult to establish, is the type of the mouthpiece, particularly since no specimen has survived from ancient Greek soil. Being made of cane, this fragile part, in contrast to some materials used for the instrument’s body, has invariably perished in the Mediterranean climate.
The primary choice anybody who ventures to reconstruct an ancient aulos faces is that between double-reed and single-reed mouthpieces. The former consists of two trapezoidal ‘blades’ fastened together to form an opening, and is nowadays used in symphonic music by bassoon and oboe players; meanwhile, a single reed, familiar from clarinets and saxophones, is formed by a single strip of cane mounted over a lateral opening on the instrument’s mouthpiece. Both reed types derive from traditional variants which not only look differently than do their symphonic equivalents but are also manufactured by entirely different procedures, which is why scholars prefer to search for the parallel aspects of ancient reed making among folk and traditional instruments that are still in use across the Mediterranean. Typical traditional single (or ‘beating’) reeds are made by cutting a ‘tongue’ from the side of a piece of cane that retains the natural closure of a node at one end. Such mouthpieces are used, for instance, on launeddas, alboka, arġūl, or miǧwiz/zummāra. Double reeds, in contrast, which are obtained by flattening the tip of a cane tube to form two parallel blades, are known for example from ciaramella, mey, duduk, ʿerāqyeh, zurna and related instruments from further East (Figure 1).5
Traditional double and single reeds. From left to right: Armenian duduk, Egyptian arġūl, Sardinian launeddas (by Pitano Perra)
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
© S. HagelThe use of a specific reed type influences the quality of sound produced by the instrument as well as its technical and tonal capacity and possible playing techniques so substantially that only one kind of reed is generally used on any type of instrument. Reed manufacturing is usually a painstaking craft on its own, and, once developed, is passed on to the next generations of players and makers. Below we will discuss written sources testifying to ancient instrumentalists adjusting certain characteristics of the reed in order to suit specific playing techniques. This clearly indicates that the musical character of an instrument depended greatly on the type of reed it used, and any change in reed design might affect the music as well as the skills that were required of the instrumentalists.
In principle, the main difference between single and double reeds in terms of playing technique regards the degree of flexibility of the material as well as the level of control over different acoustical aspects of sound exerted by the player by means of lip pressure, commonly known as ‘embouchure’. The blades of double reeds are often manufactured very soft and flexible by scraping them down after removing the harder bark layer. Controlled between the lips, they are easily compressed or expanded, allowing the player to achieve diverse sound effects such as overblowing, pitch fluctuations, vibrato and intonational adjustments. Contrarily, traditional single reeds cut out of the side of a cane tube typically do not lend themselves to lip control, but are usually more stable in terms of pitch and homogenous in terms of timbre. Additionally, other details of reed design, such as length, width, thickness and density of material greatly affect the sound in terms of pitch and intonation, volume and tone quality. Considering all these aspects of reed construction and their musical consequences, the choice of reed type is clearly paramount for fully exploring the musical potential of ancient doublepipes as well as their relations with the musical system of ancient Greece.
The significance of this choice was already noted when the first major discoveries of aulos remains, such as the pipes unearthed in Pompeii in 1867, sparked interest in the reconstruction of ancient musical instruments. The most notable pioneer in this field, Victor-Charles Mahillon, at first opted for a single reed, drawing on the analogy between the cylindrical bore of the aulos and various single-reed doublepipes of the Mediterranean, and accordingly experimented with beating reeds. Soon, however, he abandoned this idea in favour of double reeds.6 Iconographic evidence depicting doublepipes of the Pompeian type corroborated his change of mind.7 However, not all scholars were convinced that the same conclusions would necessarily apply to all known aulos specimens,8 often arguing that many different reed variants might have existed, especially considering the long history of the instrument, which spanned more than fifteen hundred years.
2 Surviving Reeds
Despite their fragility, several ancient reed fragments – though none earlier than the Hellenistic period – have withstood the destructive forces of time. The oldest one dates back to Ptolemaic Egypt and has been preserved together with one cylindrical pipe (possibly originally part of a doublepipe instrument) in the Musical Instruments Museum in Brussels (inv. 3395 and 3397).9 The reed was made from a relatively narrow stem. It features a ligatured waist just above the middle. The tip, now partly broken, was once obviously flat, but has reverted to a more cylindrical form, presumably because it was lacking a reed cap. The cane retains its bark almost everywhere, except at the very tip of the reed, where it seems to have been scraped off.
Another item that was interpreted as a fragment of a reed-mouthpiece accompanies pipe B of the so-called ‘Tibiae Gorga’ – two apparently separate aulos pipes of unknown provenience, dated approximately to the same time as the Pompeii instruments and possibly acquired by Evan Gorga at the beginning of the 20th century.10 Since the tip of the supposed reed is missing, its type cannot be confidently determined. The remaining section of the cane tube is tightly wound with a thick layer of metal wire, perhaps in order to hold it firmly in its socket on the instrument. Interestingly, a bone tube of unknown function is found in the same position on the other pipe, which calls the interpretation of the cane tube as a reed into question.11
Two other ancient reeds had survived until the 20th century, but were lost in its turbulent course. A pair of pipes with matching reeds was recovered, still in the 19th century, from a tomb located in the area of Akhmim (ancient Panopolis). One of them was described by Victor Loret in 1893, who also provided a photo of the mouthpiece.12 The image shows a badly cracked artefact that once had been a tubular piece of cane flattened towards the top. Two vertical cracks on its sides without doubt resulted from the flattening procedure. A ligature constricting the tube above the middle was added by Loret in order to prevent the fragile item from falling apart. Nevertheless, a similar binding must originally have been fastened in the same place.
Finally, a drawing in the inventory of the Egyptian Museum Berlin (inv. 12461; Figure 3) is all that is left of a double reed that once belonged to the so-called Berlin aulos, which is approximately dated to the Hellenistic or Roman period.13 The drawing depicts a cane tube with a constriction in its middle part. Between this waist and the flattened tip, the profile of the reed assumes a characteristic bulbous shape.
Suspected double reed fragments (bottom left) are rather part of wooden tubing once encased in copper alloy; note the typical twin grooves close to the left end (National Archaeological Museum of Naples, inv. 119890)
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
© S. HagelEgyptian Museum Berlin, inv. 12461 with drawing of original double-reed mouthpiece
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
© S. HagelAll attributable finds of ancient aulos mouthpieces thus represent the double-reed type. Notably, they were preserved despite their delicate blades being so much more fragile than the stout lengths of cane forming single reeds, of which – to the best of our knowledge – no ancient specimen related to the aulos survives.
3 6th–4th-Century BC Iconography and Finds
In the absence of more, and particularly of sufficiently early material evidence, we are mostly left with abundant but potentially ambiguous imagery on Greek pottery from the sixth, fifth and fourth centuries BC. Of course, iconography must be approached with due caution. Like most visual arts, the aim of Greek vase painters was not to portray reality faithfully but to reflect selective aspects through available means of conventional artistic expression. It is also important to note that many figured vessels were rather small, and the size of the painted surface usually allowed only for a very limited degree of detail and precision. Thus the size and shape of individual vases as well as the general level of accuracy need always be considered. Consequently, depictions of musical instruments are often ambiguous and have led scholars to conflicting conclusions. It is therefore imperative to analyse the largest possible number of images, looking for patterns and styles of depiction rather than focusing on single clues in isolated representations.
Even though innumerable images of aulos players exist, a majority of these do not pay particular attention to organological detail, but merely render a pipe as a couple of straight lines connected at the end. It also does not help that the instrument is most typically shown during performance – with the reeds largely hidden in the player’s mouth. More detailed depictions reveal various divisions on the instrument just below the parts that would receive the reed. These in turn may be represented either as straight or as bulging, in the latter case obviously indicating the distinctive ‘bulbs’ of Greek and Italiote doublepipe finds. One of the earliest images exhibiting divisions of the aulos tubes – a Cypriot bust from the first half of the sixth century BC – clearly shows not only bulbs (here represented as cylindrical extensions), but also two additional shorter sections that can be identified as reed stems and blades (New York 74.51.2517).14 Although the depictions of bulbs vary extremely across different art works – from a couple of almost identical oval forms to longer and shorter goblet-shaped structures – some scholars have introduced the idea of double bulbs and presented them as connected by a socket,15 wholly disregarding the fact that none of the material finds matched such a design. In an attempt to account for this perceived incongruity between iconography and archaeology, it has recently been proposed that the section previously considered the “upper bulb” actually represents the ‘stem’16 of a “long reed”.17 The argument appears to be confirmed by a splendid portrayal of the aulos depicted on a large amphora unearthed in Tarquinia (RC 6843; FIGURE 4). Both the size of the vessel and the overall meticulous craftsmanship of its decoration encourage trust in its attention to detail. Since the instrument is not represented in the act of playing, we can see its top sections in full. Apart from a somewhat elongated bulb, the pipes are furnished with a triangular reed blade, which connects to the bulb by what appears to be a slightly bulging stem. It is this section that has been identified as belonging to the reed rather than the instrument.
Attic amphora depicting a Dionysian thiasos, Museo Archeologico di Tarquinia, RC 6843
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
© K. WysluchaSuch an interpretation is supported by another depiction of doublepipes in which the reed section is marked in a different hue of ‘red’ and thus distinguished from the bulbs and the instrument’s body. This image, which comes from a tondo of a kylix attributed to the Antiphon Painter (New York 96.9.36; Figure 5), is smaller than the one depicted on the Tarquinian amphora but equally elaborate. One reed is represented from the side: what seems to be its profile resembles the bulbous form of the lost Berlin reed, terminating in a point where the blades meet. In contrast, the blade of the other reed is turned towards the viewer, revealing the horizontal edge of its tip. Both reed stems, albeit shorter and narrower, otherwise resemble the oval shape of the bulbs below them, which connect to the main body of the instrument by their ‘neck’, or – if we compare them to goblets – their foot.18 In this way, the image precisely reflects a tripartite division of the instrument’s upper end as we find it painted much more schematically in the form of apparently cylindrical sections in less detailed depictions. The kylix furthermore provides firm proof that the blades of a double reed were not only represented en face, as on amphora RC 6843 from Tarquinia, but also from other angles. The peculiar profile with a pointed tip and a bulge above the constriction, both apparently characteristic of ancient double reed blades, seem to have confused scholars who expected to find trapezoidal shapes.19
Tondo of a terracotta kylix depicting an auletes putting on a phorbeia
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
Metropolitan Museum of Art, 96.9.36. Open access3.1 Reeds Depicted From the Side
At the beginning of the twentieth century, musicologist Hermann Smith, when trying to make sense of the images of aulos reeds he had seen in the British Museum, rightly concluded that the multiple oval protrusions at the proximal ends of pipes could not have represented single reeds, but failed to recognise the sound-generating mechanism portrayed in this manner.20 A large red-figure amphora (London E271; Figure 6) that especially caught his attention depicts a profile of the double reed in a way similar to the kylix attributed to the Antiphon Painter, which we have discussed above. It shows a Muse attending to her reeds before or after a performance.21 The shape of the visible bulb is evocative of a champagne flute and thus comparable to the goblet-like design (cf. notes 36 and 37 below). A slightly shorter section corresponds to a cylindrical reed stem that leads towards the shortest section – the blade. One of the reeds terminates in a pointed tip. Its slender elongated form prompted some scholars to interpret it as a single reed,22 which is hardly conceivable, as single reeds could easily be portrayed as a mere tube deprived of any sections and protrusions. Interestingly, the other reed seems to lack a similar pointed tip. Although we cannot see all of it, its blade seems to be represented frontally, perhaps from the same angle as the one depicted on the kylix in New York. It is, thus, very likely that this image shows a double reed.
Attic amphora showing Melousa and Terpsichore
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
British Museum, E271. © The Trustees of the British MuseumThe interpretation of the characteristic pointed tip above a bulge as the tip of double reed blades finds proof in an image adorning a kylix attributed to the Ashby Painter (New York 1993.11.5; Figure 7). Conspicuously, only the extreme end of the reed is inserted into the mouth of an auletris seen in profile, so almost the entire outline of the reed is visible. Just as in the case of the depiction on the amphora in the British Museum, the reed here seems to consist of a slightly bulbous stem that is set apart from its blade by a ligature. As regards the dimensions of respective sections, the blade – terminating in the mentioned pointed tip – is shorter than the stem by approximately one third.
Exterior of an Attic kylix with a symposium scene
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
Metropolitan Museum of Art, 1993.11.5. Open accessSuch images of double reeds in profile were by no means rare. Another excellent example, found on a calyx krater (London E461; Figure 8), features an upper pipe end that consists of multiple bulbous sections. It seems that the vase suffered damage in the area next to these, whence we may be denied crucial details. Nevertheless, the three visible parts appear to embody the bulb itself, perhaps furnished with a vaguely conical reed seat, the reed stem – depicted here as the most strongly bulging section – and the reed blade seen from its distinctive profile with a pointed tip and a bulge above the constriction.23
Attic calyx-krater with a musical scene
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
British Museum, E461. © The Trustees of the British MuseumSince a piece of cane would most naturally yield cylindrical reed stems, a distinct bulge in the stem section, as is the case here and in many other instances,24 poses a considerable interpretational challenge. Which mysterious kind of procedure might have been responsible for such a shape? Is it merely a product of artistic convention, or did aulos reeds really assume this shape?25
Unfortunately, ‘long double reeds’, with or without a bulging stem, are not known from any traditional instruments still played around the world. The only specimen that could possibly serve as a parallel is a 19th-century ʿerāqyeh (a duduk-like instrument from Egypt) preserved together with its reed at the MIM in Brussels (inv. 0125).26 The upper part of its main body, starting right above the finger holes, resembles a bulb, whose upper end hosts a conical socket into which the reed is inserted. Unlike typical duduk reeds, the ʿerāqyeh mouthpiece strikingly features a constriction above a short but distinctly bulbous stem. This peculiar shape may have resulted from forcing the reed’s lower end into its seat, on the one hand, to which end some material was apparently removed from its ‘foot’, and, on the other, from the constriction below the blades. At any rate, this is presumably the only reed specimen at our disposal that possibly attests to the survival of bulging reed stems in the Mediterranean area – but on an instrument whose designation as ‘Iraqi’ points to the Middle East as its perceived place of origin.
3.2 Reed Blades Depicted en face
Fanning blades are a signature feature of the double reed, both of the modern type, manufactured by folding a gouged cane strip, and the traditional variant obtained by flattening the tip of a length of cane. Although vase paintings rarely depict this shape, the few surviving occurrences clearly point to double reeds. The most representative depiction, and as such a reference for analogous representations, is found on the previously discussed amphora from Tarquinia. A fairly similar image adorns a tondo of an early fifth-century kylix surviving in the Altes Museum Berlin (F3255). Although the painting itself is rather sketchy, both reed blades are shown quite distinctly as triangular forms, connected to the main body of the pipes by long bulging reed stems. Interestingly, the instrument lacks bulbs – the reed stems seem to be inserted directly into the main body that is slightly conical at both ends.27
Unfortunately, the majority of depictions are not equally explicit with regard to the shape and proportions of the reed blade. In many images, the blades are represented as small triangles attached to an apparent ‘upper bulb’ (cf. New York 25.78.66, 1993.11.5, London E68), while in other instances, they can hardly be told apart from conical reed seats. For example, an intriguing scene is depicted on a pelike in Münster (inv. 677).28 One of the figures – possibly a Muse – clearly adjusts her reed in preparation for a performance, holding the main body of the instrument in one hand, and in the other, two bulbous forms culminating in a flaring top, which she seems to be inspecting closely. What exactly is she eyeing so critically? Is it a bulb with a reed inserted in it, or is it rather two bulbs and a reed seat? Is the flaring top a reed blade, or perhaps a reed cap represented in this awkward way? If anything, it is more probable that the object involves a reed, as these components usually require greater attention. Thus the two bulbs could either stand for a bulging reed stem and a reed blade profile clamped at the very tip or a bulb fitted with a bulging reed stem and a short blade.29
3.3 Reed Caps/Clamps
Small objects sometimes depicted as placed across the tip of a reed almost certainly represent reed caps. Used by the players of traditional double reeds (for instance the duduk) to prevent the reeds from opening up too much, such reed caps (or ‘clamps’) have become a distinctive feature of the double reed produced in the ‘traditional’ way by flattening one end of a cylindrical length of cane. Unfortunately, of the few surviving ancient depictions that may be interpreted as representing this device several are ambiguous. One of them (Naples 80084), dating back to the second half of the fourth century BC, shows a pair of pipes linked by a piece of cord, which must be an indication of reed caps and consequently a double reed.30 Apart from the ligature, the image seems to depict the profiles of both reed blades rather roughly, tips closed by the caps that are also represented from the side. Another clear image of a piece of string linking two pipes which seem to be provided with caps was painted on a small Apulian lekythos decorated in the Gnathia technique and also dated to the fourth century BC (Melbourne NVG D17-1972). The string represented in white paint seems to be attached to the ends of the caps, in a manner similar to what is found in modern reconstructions.31
Since the caps are more recognisable when not seen from the side (where they would merely project to a small circle), several images may combine a ‘frontal’ view of caps with a side view of reed blades, most notably two probably related depictions, a well-known painting on an early fifth century lekythos and a fourth-century plate (London E583, Figure 9; Berkeley 8935).32 The former shows a satyr with two pipes in one hand and a mouthpiece in the other. The upper end of the main tubes is fitted with flame-like shapes, the interpretation of which is problematic. One of the tips is stopped by a device of a clearly horizontal rectangular shape – presumably a reed cap – while the other tip might have featured a similar device in the place where the vase has later cracked. The gesture of the satyr, who seems to be particularly focused on the reed caps, indicates that he is preparing for a performance or attending to the instrument after the performance has finished.
Attic lekythos with a satyr
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
British Museum, E583. © The Trustees of the British Museum3.4 Reed Blades, Reed Seats and Bulbs
Identifying the perishable reed in imagery implicitly entails ascribing the remainder of a depiction to the more durable instrument parts. Consequently, our endeavour to identify reeds in the iconography also requires a comparison between aulos remains from the Late Archaic and Classical period and their depictions on contemporary Greek pottery. However, apart from the visual ambiguities, the scarcity of aulos finds of the so-called ‘early type’ and their precarious dating must caution us from rushing to conclusions both regarding their potential organological variety and the lack thereof. Nevertheless, a quick overview of the archaeological material and equivalent sections depicted on vases will allow us to notice a transformation. The onset of red-figure painting gave rise to the more detailed representations of the aulos which highlight many features of the actual design regarding sections of various shapes and sizes.33 From among a group of aulos finds dated to a somewhat earlier period (first half of the sixth century), two bulb sections recovered at Perachora (A 424 and A 428)34 prove to be a meaningful point of reference for contemporary images of the instrument (cf. Figure 10).35 Their shapes closely resemble the outline of corresponding sections depicted on the amphora RC 6843 from Tarquinia (discussed above), dated to the second half of the sixth century.36 In both cases, the bulb sections consist of a thinner ‘neck’ part that extends into a wider tulip-shaped bulb, opening towards the top with a slightly flaring rim. This style of representation persists in late sixth-century works of the so-called Pioneer Group until the wake of the fifth century, with some notable examples such as a calyx krater attributed to Euphronios or Smikros (Munich 8935/9404), a stamnos attributed to Smikros (Brussels A717), and a cup attributed to Psiax (Victoria and Albert Museum, 275.64, London).37 The last image clearly distinguishes between reed and bulb by contrasting the reeds in black paint from ‘red’ bulbs and main tubes.38 The overall slender form of the doublepipes in this image foreshadows a new type of representation that seems to have emerged soon afterwards, around the turn of the century. Here both reeds and bulbs became longer and narrower, and the proportions between bulb and neck shifted towards a ‘champagne-flute’ form – a very thin neck smoothly flaring into a tulip-shaped bulb which moved from the centre of the section towards its upper part. Among images representative of this type we find many of the most famous aulos depictions from the first half of the fifth century BC, including a kylix attributed to Douris (Munich 2646), the kylix attributed to the Antiphon Painter discussed above (New York 96.9.36), a kylix attributed to Euphronios (Boston 95.27), a neck amphora attributed to the Kleophrades Painter (London E270), and many more.39 The reed section shown together with this bulb type is usually much thinner and shorter than the other parts of the instrument (cf. New York 2011.604.1.5694). Although it is difficult to determine whether the change in the style of depiction was in any way related to changes in the instrument’s design, the bulbs of auloi dated to the beginning of the fifth century BC (for instance, the Poseidonia aulos) appear to display corresponding features (cf. Figure 11).40
Images of aulos reeds and bulbs depicted on 6th- and 5th-century Greek pottery. Below the time axis, 3D models of some surviving aulos bulbs from about the same period. The bulb from Lindos survives only in a drawing. Bulbs from Ialyssos and Poseidonia are fragmentarily preserved. The latter seems to be missing the proximal rim.
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
© K. WysluchaSeveral 6th and 5th-century vase images of bulbs and reeds depicted from different angles juxtaposed with 3D models of the bulbs of Poseidonia and Perachora auloi provided with a reed reconstruction. The proportions suggested by this figure are approximate, as they do not take into account vessel curvature nor the angle of the photo.
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
© K. WysluchaObviously this type of imagery thrived along with many others that interpreted aulos bulbs and reeds in almost every possible manner – as shorter, longer, more cylindrical or more bulging. Again, such variety may have reflected an actual diversity of aulos designs for which we have limited evidence. After all, the cheapest form of doublepipes, made entirely of cane, cannot have featured any bulbs below the mouthpiece.41 The design of wooden auloi, which were probably much more common than the bone ones, eludes us. Images showing a bulb without a neck42 may be reminiscent of the wooden Elgin aulos as it is displayed in the British Museum – but it is more likely that its connection was wrongly restored, the thin-walled neck having perished as it did on the only other wooden example, a pipe from Daphne.43 Further transformations seem to have brought about the development of a more flaring reed seat, which in some cases even came to be made in a separate section from the bulb, perhaps a century later (cf. the aulos from Pydna),44 as well as a distinctive spigot at the distal end of the neck section (cf. the auloi from Taranto and Corinth). The depictions of a triangular shape above the bulb are easily mistaken for representations of a reed blade.45 Eventually, the reed seat developed into a pronounced flaring cup, clearly marking the boundary between the reed and the bulb, as can be seen in a bronze figurine from the National Archaeological Museum of Naples (cf. Figure 12). The shape of its bulbs and reed-receiving section resemble the corresponding components found for instance on the Louvre aulos, which cannot date earlier than from the Hellenistic period.
Bulbs and reed seats: bronze figurine in the National Archaeological Museum of Naples, Louvre E10962 and Taranto 12528
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
© K. WysluchaHow was the mouthpiece fitted in place? Later instruments typically feature a slightly conical socket which terminates in a distinctive step (cf. Figure 13). The shape of the socket ensures a tight seat of the reed ‘foot’ in it, while the step appears to account for the diameter of its wall, so that the cavity of the reed would continue that of the main bore as smoothly as possible. Unfortunately, the examples from early-type auloi are either insufficiently preserved – mostly broken right above the bulb where the walls become very thin – or insufficiently documented, like the bulbs from Perachora and the Copenhagen aulos. Consequently the shape of the reed-foot end and its external diameter are difficult to assess; we may only surmise that its internal diameter should have been comparatively close to that of the pipe, as any difference would have adversely affected the playable gamut. The available data from the Pydna and the Poseidonia auloi seem to suggest a value of approximately 10–12 mm for the external diameter of the reed foot (plus possibly a winding of waxed thread).46 A similar value (11.7 mm) was recorded for the internal diameter of the surviving reed socket (part Meg1
Reed sockets of Taranto 12528 and Naples 76894
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
© S. HagelThe data for all currently documented items are collected in Figure 14, comparing main bore diameters with the width of the step leading to the reed socket. Interestingly, the diagram appears to suggest some inverse correlation between the two parameters: the larger the bore, the smaller the step towards the reed socket. Since all the smallest bores come from a relatively late period, one might suspect this might be no more than a side effect of a historical development towards larger steps. However, this is not the case. On the one hand, chronology alone has no greater explanatory value than has main-bore width (32.2% vs. 32.1%: see Figure 15). On the other, the correlation becomes in fact much stronger when all late examples are excluded (R2 = 55.6%; excluding very roughly dated items: 44.3%). For the earlier subset, comprising instruments from the fourth and third centuries BC, the effect is considerable: the regression line indicates a loss of a third of a millimetre of step size per additional millimetre main bore.
Bore diameter and step towards reed socket in documented extant auloi (Pydna L/H, Taranto 12528*, Athens National Archaeological Museum BI 593, Megara 1965 L/H, Megara 1964 L/H, Dushanbe National Museum of Antiquities excav. no. 4319 and 5006, National Archaeological Museum of Naples 76892, 76893, 76894, s.n.B, 129589, Louvre E10962 L*/H*, Egyptian Museum Berlin 12461*/12462*) dark → bright: early → late *red: no reliable dating
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
Reed socket step vs. date in documented extant auloi (as in figure 14, except only roughly dated items)
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
© S. HagelPossibly the picture will change with new documented finds. The present evidence, at any rate, seems to call for an explanation. While the observed inverse relation between two dimensional variables may at first appear counterintuitive, because we might rather expect measurements to be scaled equivalently, it may, on a second glance, prove technically significant. If, as argued above, the main bore predicts the internal diameter of the reed stem, and the step is a guide to its external diameter, then we must conclude that aulos makers strove to keep the external diameter of a reed within acceptable limits, using or manufacturing thinner-walled cane tubes for instruments with large bores, while selecting more thickly walled stalks for narrower pipes. The obvious motivation for such an approach is probably to be found in the size of the blades, which is mainly determined by the cane’s external diameter. The usable tissue for reed blades lies directly below the bark, while the softer innermost layers need to be removed. Consequently, when the proximal end of the cane is perfectly flattened, the width of the blades becomes only a little smaller than half its external diameter times
3.5 Reed Length
Evidently stems that could be mistaken for additional bulbs are much longer than required for the reed’s basic functionality; physically, they operate as extensions of the main tube in its function as resonator. Apart from possible aesthetical motivations, the adoption of especially long reeds might be prompted by difficulties associated with providing a longer instrument. In the case of bone auloi, the scarcity of the material may have been decisive. Even moderately-sized instruments were typically assembled from four sections per pipe and therefore required bones from several individuals. Finding material of appropriate size was perhaps of less concern for wooden pipes, but even here, shorter bores were significantly easier to drill.
The few documented instrument remains appear to support long reeds. Even though these have not survived, a musical evaluation typically establishes a plausible total extension of the vibrating air column, which gives a good estimate for the position of the physical reed size.49 For the Pydna aulos, the only one whose total length can be established, the portion of the reed that extends above its socket thus appears to measure about 8 cm; the smaller instrument from Paestum, whose tops are broken, must have had reeds of less than 7 cm.50 On other finds, the upper part is lost or damaged to a degree that precludes even a rough guess. However, the fragments from Akanthos51 as well as the Copenhagen aulos preserve the entire main tubes, so that it becomes possible to estimate the original extension of their bulbs plus fitting reeds. If we take the total length of an instrument, i.e. the extent of its longer pipe plus its reed, as a point of reference, it turns out that in all four pairs the bulb + reed compounds measure between 27% and 34% of that total, 31.1% on average (cf. Figure 16). In comparison, in the exceptionally detailed representations shown in Figure 4–Figure 8, with eight evaluable pipes, the same value ranges between 27% and 33%, with an average of 29.9%. Evidently, Attic red-figure painters were able to portray proportions with outstanding precision.
In extant early bone auloi, instrument length is highly predictive of the combined lengths of bulb and effective reed
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
© S. Hagel4 Later Depictions
At the end of the fourth century BC the art of Greek figured vessels declined, which brought about the demise of a long tradition of depicting auloi and their reeds in specific contexts and within well-established iconographic styles. After this point, we rely on representations from a range of other art forms – most significantly, frescoes, mosaics and reliefs – across different periods and locations in the Mediterranean. Many features of the double reed design that are seen on Greek vases, such as the distinctive reed profile, triangular blades, reed caps, and a string, can also be found in such later artworks.52 This suggests that the aulos continued to be sounded with double reeds throughout antiquity, even if the shape of both reed and instrument underwent a series of changes. While the finds of post-classical auloi clearly attest to an evolution of instrument design, we have much less evidence for the changes of the reed itself. It is very probable that the long stem below the reed’s constriction had been considerably shortened, up to the proportions that are familiar from the drawing of the specimen held in the Egyptian Museum Berlin. Late fourth- century evidence – for instance, images of double pipes painted on Gnathian pottery53 – suggests that the reeds of certain aulos types might have been already shorter in this period.
Few aulos representations survive from the Hellenistic era (third to first century BC). This prevents us from observing any structural transformation that aulos reeds might have undergone in this time. We encounter the next large group of pictorial evidence on Pompeian frescoes, most significantly the famous painting from the Inn of the Sulpicii in Moragine.54 The painter of this detailed image of double pipes with metal mechanisms clearly marked the triangular device in the right pipe in an off-yellow hue, which could suggest a reed rather than a cup-like reed seat, which often assumes a similar shape. However, the apparent absence of a tubular reed stem or a conical seat between the bulb and triangle calls this interpretation into question, as does the shading of the element which appears incompatible with a flattened object. If the painting indeed shows a reed, then its depiction fails to render the crucial details of the instrument’s design in a realistic manner.55 Unfortunately, the paint faded away in the place of the other reed, preventing comparison. In this case the reed seat is either very small or entirely absent, as the long triangular reed blade seems to emerge directly from the bulb. There is also no sign of any reed stem, long or short.
Quite a different set of reeds is depicted on a mosaic held at the Capitoline Museum in Rome, dated to the second century AD. This image inspired Mahillon’s change of mind with regard to the type of aulos reed, as it clearly shows the reeds closed with caps – a signature feature of the double reed (cf. n. 6).56 The blades seem more elongated, with almost parallel sides in contrast to the trapezoidal and triangular shapes familiar from the representations discussed above. Below the constriction, we discern a slightly bulging reed stem, whose visible part is considerably shorter than the blades.
One of the most famous double-reed depictions, which is at the same time related to the so-called ‘Phrygian’ aulos, is a Roman relief showing an archigallus of Cybele surrounded with cultic paraphernalia (inv. 1207), probably found at Lanuvium, now held at the Musei Capitolini and also dated to the second century AD. This three dimensional representation leaves not the slightest doubt about the nature of the reed in one of the pipes; its characteristic trapezoidal blades are fastened together just above the topmost rim of the instrument. The presence of a bulb furnished with a flaring reed seat supports this identification.57 The size of the reed blades approximately corresponds to the size of the latter, suggesting that the reed stem hidden inside this section might have been nearly equal in terms of length. Based on this splendidly detailed representation, we may wonder whether or not similar depictions of the Phrygian pipes in a votive context also include reeds. For instance, two bas-reliefs from the taurobolic altar of Scipio Orfitus dated to the fourth century AD as well as another bas-relief perhaps coming from the Roman Phrygianum show Phrygian auloi with intersecting pipes or suspended from the arbor sacra, provided with reed-like shapes emerging from proximal pipe ends.58 However, these structures are difficult to tell from reed seats due to the less detailed style of representation, so that an interpretation as reeds ultimately depends only on the comparison with the Lanuvium relief.
In the final centuries of the Western Roman Empire, numerous images of auloi are found in connection with Euterpe – one of the nine Muses, whose popularity as the topic of mosaics and sarcophagus reliefs was especially widespread at that time.59 And although the majority of these representations portray reeds rather carelessly, if at all, several of the more detailed depictions yield interesting clues. Most importantly, an intriguing scene from a mosaic from the fourth century AD found at the Villa of Arellano in Spanish Navarra shows Euterpe in the company of Hyagnis, the mythical inventor of aulos playing, apparently comparing an (older?) plain aulos type to (a modern?) one equipped with intricate mechanisms including metal side-tubes.60 Both pipes of this set appear provided with double reeds, distinguished not only by the characteristic trapezoidal shape of their blades but also by a bright hue of yellow that contrasts with the darker shades used for the metal components of the instruments. The reed blades, which appear to be inserted directly into the bulbs, seem large in proportion to the body of the pipes, but the representation otherwise does not inspire particularly great trust in its accuracy. The gesture with which Hyagnis attends to the mouthpiece of his pipe suggests that it is indeed a reed, so that similar structures at the proximal ends of Euterpe’s pipes should also be viewed as reeds.
In comparison with the depictions of aulos mouthpieces on Greek figured pottery, later images seem generally to portray double reeds en face, so to speak, displaying their characteristic triangular/trapezoidal blade shape. Meanwhile, views from other angles, exhibiting the bulging, are almost entirely absent from Roman-era aulos painting (as opposed to three-dimensional representations). Moreover, images from that time seem to attest to structural transformations that the aulos reed might have undergone since the Classical period: it seems to have grown smaller, while its stem was shortened, extending from the instrument no further than approximately the length of the blade (cf. the reconstructions in Figure 17). This conclusion largely coincides with the results of reed length calculations based on the aulos finds from the Roman Imperial era. For instance, the estimated effective reed extensions for the four Pompeii pipes (National Archaeological Museum of Naples, inv. 76891–4) range between 30 and 45 mm, compared to 68.5–79.5 mm estimated for the aulos from Paestum.61
Double reeds. From left to right: duduk reed, historical bassoon reed, reconstructed aulos reeds of various lengths
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
© K. Wyslucha5 Literary Testimonies
While the iconography clearly favours double reeds, up to suggesting a development in their proportions, we need to remain aware that this type of evidence is inherently biased: we can only expect to find a subset of ancient instrument types represented, which most likely included those played by an elite or otherwise of high cultural esteem or of important mythological or symbolic value. Similar considerations may in principle affect any kind of evidence. However, while the existence of particularly favoured instrument types in visual representations may easily eclipse all others, texts are much more flexible in this respect, especially when instruments are not the primary focus but invoked for extraneous reasons. A study of ancient reeds must therefore remain incomplete without looking at some important literary passages. In general, these confirm the picture of an auletic culture dominated by double reeds; however, we will see that the single reed was also known.
The most notorious text on our topic, Theophrastus’ account of the nature and use of cane growing in Lake Copais in Boeotia (Hist. Plant. 4.11.1–9), remains tantalisingly silent on the manufacture of the mouthpieces once the stalks are prepared and cut down to internodia. Even so, it has been argued that his terminology is consistent with or indeed favours a double-reed interpretation.62 He employs two different terms for the mouthpieces, glôtta (“tongue”) when referring to an individual reed, and the plural zeúgē (literally, “yokes”) when viewing the objects more generally. Both must have been in common use in Theophrastus’ time; the latter had spawned derivatives such as zeugítēs, denoting the kind of cane suitable for mouthpieces (3; 5), as well as zeugopoiía, the craft of their makers (6). While glôtta is unspecific because it may denote anything that enables something to speak, zeûgos undoubtedly denotes some kind of physically paired items and was therefore taken to refer to “not the separate mouthpieces of a pair of auloi, but the twin blades of the double-reed mouthpiece” (West 1992, 84).63 However, this conception might work better for modern mouthpieces, which are indeed joined from two separate blades just as a pair of separate oxen is joined under the yoke. In contrast, ancient manufacture by flattening an integral length of cane does not suggest an idea of joining. On the other hand, some depictions clearly show the two reeds of an aulos connected by a kind of thread (fastened to their protective caps),64 which would keep the matching pairs together when being detached from the instrument. In this way it appears more likely that a zeûgos would have referred to a pair of reeds, i.e. the smallest item that a *zeugopoiós would have sold. Unfortunately we seem never to encounter the term in the singular, while the attested instances work equally well when read as referring to either pluralities of reeds or pluralities of reed pairs (4
Later medical texts confirm the double nature of contemporary aulos reeds, as are often clearly discernible in Roman-period iconography, by associating the shape and function of the vocal cords with them. Galen expressly designates the vocal cords by the same variation of tongue-related terminology that was applied to reeds, glôtta and glōttís, talking as if he was creating the terminology:
τουτὶ τὸ σῶµα τὸ προκείµενον ἐν τῷ λόγῳ νῦν, ὃ δὴ γλωττίδα τε καὶ γλῶτταν ὀνοµάζω λάρυγγος .
the part on which we are now focussing, which I therefore call the glōttís and glôtta of the larynx.
GALEN. De usu partium, 3.562.13–15
While this passage might possibly be dismissed as figurative, the more elaborate reasoning of Oribasius clearly presupposes double reeds:
ἀλλ᾿ ἐπεὶ δέδεικται πρόσθεν ἀναγκαῖον ὑπάρχειν εἰς γένεσιν φωνῆς αὐξηθῆναι τῆς φορᾶς τοῦ πνεύµατος τὸ τάχος, αὐξάνεσθαι δ᾿ ἐδείκνυτο κατὰ τὰς στενοτέρας τῶν ὁδῶν, εὐλόγως ἡ φύσις ἔνδον τοῦ λάρυγγος εἰργάσατο τοιοῦτον ἀκριβῶς ὄργανον οἷόνπερ ἐν τοῖς αὐλοῖς ἐστιν ἡ γλωττίς· τὸ κάτω µὲν γὰρ αὐτῆς πέρας ἀκριβῶς ἐστι στενόν, ἄνω δ᾿ εὐρύνεται πρὸς τὴν τῆς φωνῆς γένεσιν ἐπιτηδείως· ἐπίκειται δ᾿ ἐφεξῆς ἄνωθεν µὲν τῆς γλωττίδος τὸ πέρας τοῦ λάρυγγος, ἐφεξῆς δ᾿ ἡ φάρυγξ, εἶθ᾿ ἑξῆς δίκην ἠχείου τινὸς ὁ τοῦ στόµατος οὐρανός .
Since it was shown above that in order to produce sound it is necessary to increase the velocity of the breath, and also that such an increase is brought about wherever its path narrows down, it is according to good reason that inside the larynx nature has realised exactly such a tool as forms the glōttís on the auloi. For its lower end is exactly narrow, while it expands higher up, as is useful for the production of sound. Immediately above, the top of the larynx abuts on the glōttís, then the pharynx, then, like a kind of resonator, the palate.
ORIBAS. Coll. med. 62.27
In Oribasius’ physics, air velocity is fundamental for sound production, and he correctly considers that the velocity of a substance must increase when being coerced through a narrow passage. The vocal cords provide precisely such a narrow passage, subsequently expanding into the upper part of the larynx and the pharynx. It is in this combination that Oribasius finds a precise equivalent to the aulos reed. Figure 18 shows the glottis in action (a view that was hardly available to the ancient doctor) next to the opening of a reconstructed aulos mouthpiece; indeed both work in very similar ways. Equally importantly, the side view of the aulos reed illustrates the expansion from a narrow slit to ultimately the diameter of the instrument’s resonator, which Oribasius considers a functional requirement for convenient sound production.
Vocal cords and double reeds. Images: upper left: James P. Thomas, https://www.voicedoctor.net/media/normal-vocal-cord/normal-female-vocal-cords/normal-female-vocal-cord/ (Creative Commons Attribution-NonCommercial-ShareAlike 3.0 unported license); upper right, lower: reconstructions of Roman-period aulos reeds
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
© S. HagelWithin this general semblance, Galen identifies another functional parallel: both the human and the aulos glottis require moistening:
ὑγρὸν δ᾿ οὐχ ἁπλῶς, ἀλλὰ σὺν τῷ γλίσχρον τέ πως εἶναι καὶ λιπαρόν, ἵν᾿ ἐπιτέγγηται διὰ παντὸς οἰκείᾳ νοτίδι καὶ µή, καθάπερ αἱ τῶν αὐλῶν γλῶτται ξηραινόµεναι συνεχῶς ἐπικτήτου τινὸς ὑγρότητος δέονται, καὶ αὐτὸ τῶν ἔξωθεν ἰαµάτων ἐπιδεὲς γίγνηται .
but not simply moist, but with a greasy and oily quality, so that [the glottis] is wholly humidified by a moisture of its own and does not (such as do aulos tongues, which dry out and constantly need some additional supply of humidity) come to require external remedy.
GALEN. De usu partium, 3.566.8–12
Moistening is a typical procedure with double reeds, both those of Western orchestra instruments and traditional ones such as the duduk. The blades of a duduk reed close up when drying, and so do many reconstructed aulos reeds, rendering the pipe unplayable instead of only compromising the sound. Traditional single reeds, in contrast, are never watered and their general sound is little affected by humidity.
Galen’s reasoning appears to stand in a tradition that reached back several centuries. The Peripatetic “On things heard” not only refers to the same experience but also considers a potential cause:
δεῖ δὲ καὶ τῶν αὐλῶν εἶναι τὰς γλώττας πυκνὰς καὶ λείας καὶ ὁµαλάς, ὅπως ἂν καὶ τὸ πνεῦµα διαπορεύηται δι᾿ αὐτῶν λεῖον καὶ ὁµαλὸν καὶ µὴ διεσπασµένον. διὸ καὶ τὰ βεβρεγµένα τῶν ζευγῶν καὶ τὰ πεπωκότα τὸ ⟨ν ?〉σίαλον εὐφωνότερα γίγνεται, τὰ δὲ ξηρὰ κακόφωνα .
Similarly it is necessary that the tongues of the auloi are dense, smooth and even, so that the breath can also travel through them smoothly and evenly and does not become dispersed. This is the reason why moistened mouthpieces and those that are imbued with saliva sound better, but dry ones worse.
[ARISTOT.] De audib. 802b
It is of particular interest that two ways of moistening are mentioned, which seem to refer to (a) soaking before playing, presumably in water (
The particular association of aulos reed moistening with the human vocal organ appears to take literary attestation of double reeds back to the late Classical Period. Other parallels exploited in the same text provide more circumstantial evidence in the same direction:
παχεῖαι δ᾿ εἰσὶ τῶν φωνῶν τοὐναντίον, ὅταν ᾖ τὸ πνεῦµα πολὺ καὶ ἀθρόον ἐκπῖπτον· διὸ καὶ τῶν ἀνδρῶν εἰσὶ παχύτεραι καὶ τῶν τελείων αὐλῶν, καὶ µᾶλλον ὅταν πληρώσῃ τις αὐτοὺς τοῦ πνεύµατος. φανερὸν δ᾿ ἐστίν· καὶ γὰρ ἂν πιέσῃ τις τὰ ζεύγη, µᾶλλον ὀξυτέρα ἡ φωνὴ γίγνεται καὶ λεπτοτέρα, κἂν κατασπάσῃ τις τὰς σύριγγας .
Contrarily, voices are thick whenever the breath emanates in abundance and densely. This is the reason why those of males are thicker, and those of ‘complete’ auloi, and even more if one fills them with breath. This is obvious: for also if one compresses the mouthpieces, the voice becomes higher and feebler, and also if one draws down the sýringes.
[ARISTOT.] De audib. 804a
The so-called sýrinx (a term whose basic meaning is something like ‘whistle’), a small hole close to the upper end of a pipe, normally closed by a plug or some kind of mechanism, raised the pitch substantially by at least tripling its frequency.66 Compressing the reed, in turn, might be conceived as another way of achieving the same effect of ‘overblowing’ to a higher harmonic mode. However, according to our experience, reconstructed aulos mouthpieces rather overblow when compressed less. More probably, the text refers to the much more delicate pitch adjustment, in a microtonal range, effected by decreasing the volume of the reed cavity. At any rate, both effects are only possible with double reeds; a traditional single reed would only close up when being compressed, muting the instrument.
6 Single Reeds
Pressing the reed blades is also mentioned in an earlier passage in the same work, with focus not on pitch change but a harder, brighter sound. At the same time, it appears to form the only testimony to the use of single reeds in Greek antiquity:
δῆλον δ᾿ ἐστὶ κἀπὶ τῶν αὐλῶν. τὰ γὰρ ἔχοντα τῶν 〈ζ 〉ευ 〈γ 〉ῶν 67τὰς γλώττας πλαγίας µαλακωτέραν µὲν ἀποδίδωσι τὴν φωνήν, οὐχ ὁµοίως δὲ λαµπράν· τὸ γὰρ πνεῦµα φερόµενον εὐθέως εἰς εὐρυχωρίαν ἐµπίπτει, καὶ οὐκέτι φέρεται σύντονον οὐδὲ συνεστηκός, ἀλλὰ διεσκεδασµένον. ἐν δὲ ταῖς συγκροτ 〈ητικ 〉αῖς γλώτταις ἡ φωνὴ γίνεται σκληροτέρα καὶ λαµπροτέρα ἂν πιέσῃ τις αὐτὰς µᾶλλον τοῖς χείλεσι, διὰ τὸ φέρεσθαι τὸ πνεῦµα βιαιότερον .
This also becomes clear from the auloi. For the mouthpieces that have their tongues sideways produce a sound that is softer, though not similarly bright. For as the breath moves, it falls right into a wide space and no longer moves in a tight and dense condition, but becomes dispersed. In the strike-together-style tongues, on the other hand, the sound becomes harder and brighter if one exerts greater lip pressure on them, because the breath moves more vigorously.
[ARISTOT.] De audib. 801b
The argument associates the acoustic quality of ‘brightness’ with the idea of a condensed regular stream of air. This is in principle provided by a kind of reed ‘tongues’ that is described with some form of synkrot-, ‘striking together’; the precise word form may have fallen victim to a careless copyist. Such a description points to double reeds, the shape of which elucidates the author’s reasoning: the narrow opening guarantees precisely the required condensed jet of air, which the gradual spreading of the blades subsequently safeguards from sudden dispersal (cf. Figure 19). On the other hand, the “mouthpieces that have their tongues sideways” must be the well-known type of single reeds cut from the wall of the cane tube.68 Here the air, after passing around any of the three edges of the rectangular tongue, indeed “falls right into a wide space”: the internal cavity of the reed retains much the same diameter from its foot up to its top. The sound qualities the author associates with either type of reeds appear to confirm such a reading: single and double reeds typically align themselves in this way along a perceptional axis between softness and brightness, in both modern orchestra and traditional instruments.69
Conceived airflow in single and double reeds according to [Aristot.] De audib. 801b
Citation: Greek and Roman Musical Studies 11, 2 (2023) ; 10.1163/22129758-bja10062
© S. HagelIt emerges that single reeds were part of the ancient Greek soundscape, after all. However, since all attributable depictions portray instruments equipped with double reeds, and almost all relevant textual passages refer to the same kind, we must suspect that single reeds did not belong to the cultural contexts typically represented in elite discourse. We might search for them mainly in non-urban contexts. Perhaps we can identify one instrument type of widespread usage, at least in the Hellenistic period; but that will be a different story.
7 How to Use Double Reeds
What are the advantages of double reeds? Above all, they lend themselves to advanced control both by the lips of the player, between which their blades vibrate, and by her tongue, whose tip may touch the ends of either or both reed tips of a pair. Tonguing allows sharp onsets and stops of individual notes, separating them in the way of staccato playing or by rests. Both techniques appear evidenced by later notational practice, in contexts that may be associated with the auletic art at least indirectly.70
Since the Peripatetic author was aware of the effects of changing lip pressure, it is likely that these were also exploited by musicians. Theophrastus in turn grants insights into a music-historical trade-off between mutually exclusive properties of reeds:71
Τὴν δὲ τοµὴν ὡραίαν εἶναι πρὸ Ἀντιγενίδου µέν, ἡνίκ᾿ ηὔλουν ἀπλάστως, ὑπ᾿ Ἄρκτουρον Βοηδροµιῶνος µηνός· τὸν γὰρ οὕτω τµηθέντα συχνοῖς µὲν ἔτεσιν ὕστερον γίνεσθαι χρήσιµον καὶ προκαταυλήσεως δεῖσθαι πολλῆς, συµµύειν δὲ τὸ στόµα τῶν γλωττῶν, ὃ πρὸς τὴν διατορίαν εἶναι χρήσιµον. ἐπεὶ δὲ εἰς τὴν πλάσιν µετέβησαν, καὶ ἡ τοµὴ µετεκινήθη· τέµνουσι γὰρ δὴ νῦν τοῦ Σκιρροφοριῶνος καὶ Ἑκατοµβαιῶνος ὥσπερ πρὸ τροπῶν µικρὸν ἢ ὑπὸ τροπάς. γίνεσθαι δέ φασι τρίενόν τε χρήσιµον καὶ καταυλήσεως βραχείας δεῖσθαι καὶ κατασπάσµατα τὰς γλώττας ἴσχειν· τοῦτο δὲ ἀναγκαῖον τοῖς µετὰ πλάσµατος αὐλοῦσι .
They say, before the time of Antigenidas, when they played without fashioning (aplástōs), the time for cutting had been under Arcturus in the month of Boedromion. If cut in this way, it becomes useable within several years and requires much preliminary playing, but the mouth of the tongues closes up, which is useful for a piercing sound (diatoría).72 When they had embraced the fashioned style (plásis), the cutting had changed too: now they cut it during the months of Skirrophorion and Hekatombaion, shortly before or just at solstice. They say it becomes useful within three years, requires little preliminary playing, and the tongues can do73 kataspásmata. This is necessary for those playing with fashioning (plásis).
THEOPHR. Hist. plant. 4.11.4–5
Switching to a significantly earlier harvesting season would almost certainly reveal a demand for softer material.74 Harder reeds from before the time of Antigenidas, presumably in the first half of the fourth century BC, warranted a brighter sound, which the author associates with more closed reeds – perhaps following precisely the same physical ideas as we found above expressed in a dedicated acoustical treatise from the same philosophical school. Accordingly, the later softer reeds may generally have been more wide open. Consequently, a brighter sound would have been achieved by narrowing down their ‘mouths’ using lip pressure.
The modern style is associated with the terms plásis and kataspásmata, the latter being a prerequisite for the former. The idea of kataspân, ‘pulling downwards’, is elsewhere found associated with downwards pitch movement: Antyllus uses the term when recommending starting vocal exercises in the lowest region in the voice.75 Similarly, Sextus Empiricus describes deviations from a defined pitch in either direction by the notion of ‘pulling’.76 Both examples come from a significantly later period, when the spatial metaphor of pitch movement had more firmly established itself; but it is hard to see what Theophrastus’ kataspásmata might be other than a technique for decreasing the pitch of a note while it is sounded. But decreasing by what extent? Would the pitch change have involved closing finger holes, resulting in abrupt intervallic jumps, perhaps a technique of ‘closed fingering’, as argued by Maurice Byrne?77 Or would the pitch changes have been tiny, effected either by pulling the reed a few millimetres further out of the mouth, so that the lips would come to rest closer to its tip, or by reducing lip pressure, reversing the procedure of increasing the pitch by increased lip pressure described above. The image of ‘pulling’ or ‘drawing’ would rather suggest the latter techniques, both of which have a similar effect of ‘bending’ the pitch, which also accords with Pliny’s rendition of Theophrastus’ text as ad flectendos sonos (NH 16.171).78
It is above all the modification of lip pressure that evokes an idea of ‘shaping’ as underlies the term plásis. Imagining Antigenidas (and his followers) sort of ‘shaping’ each note with the lips also adds a much deeper meaning to his characterisation by Apuleius as omnis uoculae melleus modulator, “honey-sweet performer of even the least significant note” (Florida 4).
This nicely coherent image is however cast into doubt by other passages, above all one from Quintilian’s instructions on rhetorical education, where the idea of plas- appears associated with aulos-playing in a very different way:
Ne illas quidem circa s litteram delicias hic magister feret, nec uerba in faucibus patietur audiri nec oris inanitate resonare nec, quod minime sermoni puro conueniat, simplicem uocis naturam pleniore quodam sono circumliniri, quod Graeci catapeplasmenon dicunt [sic appellatur cantus tibiarum quae praeclusis quibus clarescunt foraminibus recto modo exitu grauiorem spiritum reddunt].
Such a teacher will not even bear that affected treatment of the letter ‘s’, nor will he accept that words are pronounced in the back of the mouth or resound in its empty space, nor – what least befits a pure mode of speech – that the straightforward nature of the voice is adorned with a kind of fuller sound (what the Greeks call ‘katapeplasmenon’ [this term denotes the playing of tibiae which, when the openings through which they emit a bright sound are closed, yield a lower breath by the straight passageway alone]).
QUINT. Inst. or. 1.11.6–8
The passage is problematic in several ways. Firstly, the explanation of the term catapeplasmenon interrupts the argument without contributing anything to it. One might perhaps expect a further explanation on how this technique sounded when applied in speech, but what is given refers only to the aulos and can hardly be applied analogously. Secondly, the interruption is not marked by “sic enim”, as we would expect within a continuous text.79 Thirdly, it makes little sense as such. It appears to describe the bass note of an aulos with all side holes closed, a note that may well have a particular sound, but none that required a special playing technique that might reasonably have been designated as catapeplasmenon. The sentence therefore bears all hallmarks of a marginal note that some copyist mistook for part of the text. This suspicion is confirmed by the paraphrase of the passage in Julius Victor’s Ars rhetorica (p. 441 Halm), where the dubious sentence is omitted even though the surrounding phrases are repeated verbatim. While Quintilian obviously could count on at least some of his readers’ familiarity with the Greek term, it would have been opaque to a later period. Whoever added the ‘explanation’ appears to have used a source that discussed aulos playing, not rhetoric. We cannot know what this source had originally stated: did it talk about plásis or was there indeed an auletic interpretation of katapeplasménon as well? Taken at face value, it doubtless appears to associate some form of plas- with the employment of bass notes, and has consequently prompted Byrne’s interpretation in terms of ‘closed fingering’. On the other hand, it does not as such describe any of these techniques; the image of sustained bass notes it literally invokes is not qualified by a notion of ‘frequently’ or ‘intermittently’ as we would expect.
At any rate, the context in Quintilian forms no reliable basis for interpreting the original sense of auletic plásis either. The weight of the preposition should not be underestimated; while plásis may mean any kind of shaping, katapeplasménos more typically refers to make-up (including three-dimensional applications) or making up. Both forms have been applied to speech. Quintilian himself had already referred to plasma as an effeminate way of talking – once more presuming his readers would be familiar with the term:80
Sit autem in primis lectio uirilis et cum sanctitate quadam grauis, et non quidem prorsae similis, quia et carmen est et se poetae canere testantur, non tamen in canticum dissoluta nec plasmate, ut nunc a plerisque fit, effeminata.
Above all, poems are to be read in a manly manner, with a kind of holy dignity, certainly unlike prose, because it is ‘song’ and the poets attest to being ‘singers’, but also not dissolved into a tune, and not, as is nowadays usual, effeminated by plasma.
QUINT. Inst. or. 1.8.2
In a medical context, on the other hand, reciting with plásma could be regarded a particularly useful exercise:
καὶ περιπάτοις ἀναφωνήσεσίν τε καὶ ἀναγνώσεσιν µετὰ πλάσµατος καὶ γυµνασίοις τοῖς δι᾿ ὀρχήσεως κωρύκου τε .
by walking, voice exercises and reading with plásma, exercises through dancing, punch-balls.
SORANUS Gynaec. 1.49
Particularly illuminating, finally, is the appearance of the term in a very technical context in Aristides Quintilianus, regarding the various combinations of melody, text and rhythm as the components of music:
ῥυθµὸς δὲ καθ᾿ αὑτὸν µὲν ἐπὶ ψιλῆς ὀρχήσεως, µετὰ δὲ µέλους ἐν κώλοις, µετὰ δὲ λέξεως µόνης ἐπὶ τῶν ποιηµάτων µετὰ πεπλασµένης ὑποκρίσεως, οἷον τῶν Σωτάδου καί τινων τοιούτων .
Rhythm [appears]: on its own, in unaccompanied dance; together with melody, in kola; combined solely with diction, in poems with fashioned (peplasménē) performance, such as those of Sotades and others of this kind.
Aristid. Quint. Mus. 1.13, 31.28–32.2 W.-I.
Obviously, plásis is here inextricably connected with whatever contributes rhythm to what would otherwise be mere recitation, sporting metre but no rhythm (as seems to be presupposed for all other kinds of poetry that was not delivered as genuine song). Sotadeans are an extremely flexible metre combining trochaic and iambic elements within an overall ionic framework. While more regular forms, above all hexameters, elegiac distichs, and iambic verse convey their metrical form to anybody versed in the tradition even when enunciated without particular attention to keeping pace with an abstract rhythm, the same cannot be assumed for Sotadeans. Instead, performers would conceivably have adopted other means of expressing the rhythm. Whatever these may have been in detail – we need at least to expect special attention to a more accurate expression of syllable length, combined with some kind of non-pitched inflection expressing the structure in terms of thésis and ársis – it is little wonder that in ordinary speech or oration similar vocal techniques would have been regarded as artificial, and consequently judged as expressions of affectation.
All this does not necessarily narrow down the potential meanings of the same term in the inherently musical art of the aulos, since both the rhetoric and the auletic usage may have developed independently from the general idea of ‘artificially shaping’; perhaps the former was also loosely derived from the latter. Quintilian’s expression pleniore quodam sono circumliniri, “adorn with some fuller kind of sound”, may adequately express the common nucleus. If the shared designation rested on any more technical ground, we would have to look to ‘moulding’ the sounds using not only the lips but the entire configuration of palate, tongue and pharynx, the shape of which determines the sounds of the various vowels in human speech by enforcing and suppressing different harmonics of the vocal cords’ basic frequency. Similarly, it is possible to influence the colour of an aulos note, since the oscillation of the reed blades is governed by both resonator cavities to which they are coupled, not solely the pipe but also the mouth cavity. In this way, the concept of plásis would acquire an overarching technical meaning. While controlling the opening of the reed would doubtless still be part of the picture, we would have to add even more refined techniques producing subtle nuances of instrumental sound. A soft reed might be essential for these as well, as it would more easily subject its oscillations to the harmonic configurations suggested by the resonators. But this can be little more than speculation, perhaps endorsed by the prestige Antigenidas’ innovation enjoyed over many centuries. Consequently we must stop here, handing the argument over to experimental auletics.
Acknowledgements
We greatly appreciate the generosity of Stelios Psaroudakēs, Christina Majnero and Roberto Stanco in sharing unpublished research data.
Parts of this publication are based on research funded by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No. 787522), respectively. The views presented here, however, reflect only those of the authors; the ERCEA is not responsible for any use that may be made of the information contained.
Bibliography
Amigues, S. (1989). Theophraste, Recherches Sur Les Plantes: Tome II: Livres III–IV. Paris: Les Belles Lettres.
Baines, A. (1967). Woodwind Instruments and Their History. New York: Dover.
Banou, O. (1998). Ο δίαυλος τής Πύδνας . The Double Flute Found in Pydna. In: Αρχαία ελλενική τεχνολογία. Πρακτικά του 1ου διεθνούς συνεδρίου, Θεσσαλονίκη, 4–7 Σεπτεµβρίου 1997 [Archaia hellenikē technologia. Praktika tou 1. diethnous synedriou, Thessalonikē, 4–7 Septembriou 1997], Thessaloniki: Εταιρεία Μακεδονικών Σπουδών [Etaireia Makedonikōn Spoudōn], pp. 519–524.
Barker, A.D. (1984). Greek Musical Writings, Volume 1: The Musician and His Art. Cambridge et al.: Cambridge University Press.
Barker, A.D. (1989). Greek Musical Writings, Volume 2: Harmonic and Acoustic Theory. Cambridge/New York: Cambridge University Press.
Barker, A.D. (2015). Porphyry’s Commentary on Ptolemy’s Harmonics. Cambridge: Cambridge University Press.
Bartholin, C. (1679). De tibiis veterum et earum antiquo usu: libri tres (2nd ed.). Amsterdam: Wetstenium.
Becker, H. (1966). Zur Entwicklungsgeschichte der antiken und mittelalterlichen Rohrblattinstrumente. Hamburg: Musikverlag H. Sikorski.
Bélis, A. (1986). L’aulos Phrygien. RA 1986, pp. 21–40.
Bélis, A. (1988). Studying and Dating Greek Auloi and Roman Tibiae: A Methodology. In: E. Hickmann and D.W. Hughes, eds, The Archaeology of Early Music Cultures. Third International Meeting of the ICTM Study Group on Music Archaeology, Bonn: Verlag für systematische Musikwissenschaft, pp. 233–248.
Bellia, A., and Pavone, D.P. (2021). Computed Tomography and Handcrafting Processes of an Ancient Musical Instrument: The Aulos from Poseidonia. Archeologia e Calcolatori 32, pp. 375–401. DOI: 10.19282/ac.32.1.2021.21.
Blinkenberg, Ch. (1931). Lindos. Fouilles de l’Acropole 1902–1914, Vol. I: Les Petits Objets. Berlin: De Gruyter.
Bundrick, S.D. (2005). Music and Image in Classical Athens. New York, NY: Cambridge University Press.
Byrne, M. (2000). Understanding the Aulos. Studien zur Musikarchäologie (Orient-Archäologie) 2 (7), pp. 279–285.
Byrne, M. (2002). Understanding the Aulos II. Extended Pipes and Drone. Studien zur Musikarchäologie (Orient-Archäologie) 3 (10), pp. 367–373.
Cervelli, L., ed. (1994). La Galleria Armonica: Catalogo del Museo degli Strumenti Musicali di Roma. Roma: Istituto Poligrafico e Zecca dello Stato.
Chailley, J. (1979). La Musique Grecque Antique. Paris: Belles Lettres.
Closson, E. (1930). Une nouvelle série de hautbois égyptiens antiques. In: Studien zur Musikgeschichte. Festschrift für Guido Adler, Wien-Leipzig: Universal, pp. 17–25.
Curtis, J. (1914). The Double Flutes. JHS 34, pp. 89–105. DOI: 10.2307/624480.
Düring, I. (1934). Ptolemaios und Porphyrios über die Musik. Göteborg: Elander.
Goulaki-Voutyra, A. (2020). Playing or Not Playing the Auloi: How to Read the Images on Attic Vases. Series Musicologica Balcanica, Vol. 1, pp. 358–383. DOI: 10.26262/ SMB.V1I1.7765.
Hagel, S. (2008). Re-Evaluating the Pompeii Auloi. JHS 128, pp. 52–71.
Hagel, S. (2010). Understanding the Aulos Berlin Egyptian Museum 12461/12462. Studien zur Musikarchäologie (Orient-Archäologie) 7 (25), pp. 67–87.
Hagel, S. (2012a). The Pompeii Auloi: Improved Data and a Hitherto Unknown Mechanism. Studien zur Musikarchäologie (Orient-Archäologie) 8 (27), pp. 103–114.
Hagel, S. (2012b). The Aulos Sýrinx. Rudiae 22–23, pp. 491–518.
Hagel, S. (2018). ‘Musics’, Bellermann’s Anonymi, and the Art of the Aulos. GRMS 6.1, pp. 128–176. DOI: 10.1163/22129758-12341316.
Hagel, S. (2019). Reconstructing the Auloi from Queen Amanishakheto’s Pyramid. Studien zur Musikarchäologie (Orient-Archäologie) 11 (40), pp. 177–197.
Hagel, S. (2020). Understanding Early Auloi: Instruments from Paestum, Pydna and Elsewhere. In: G. Zuchtriegel and A. Meriani, eds, La tomba del Tuffatore: rito, arte e poesia a Paestum e nel Mediterraneo d’epoca tardo-arcaica, Pisa: ETS, pp. 421–459.
Hagel, S. (2021). Assessing Unknown Parameters of Instrument Finds by Writing Software. Archeologia e Calcolatori 32, pp. 403–421. DOI: 10.19282/ac.32.1.2021.22.
Howard, A.A. (1893). The Αὐλός or Tibia. HSPh 4, pp. 1–60. DOI: 10.2307/310399.
Howard, A.A. (1899). The Mouth-Piece of the Αὐλός. HSPh 10, pp. 19–22. DOI: 10.2307/310294.
Landels, J.G. (1964). Fragments of Auloi Found in the Athenian Agora. Hesperia 33, pp. 392–400.
Landels, J.G. (1999). Music in Ancient Greece and Rome. London/New York: Routledge.
Loret, V. (1893). Sur une ancienne flute égyptienne découverte dans les ruines de Panopolis. Publications de la Société Linnéenne de Lyon 12, pp. 91–101. DOI: 10.3406/linly.1893.12036.
Loret, V. (1913). Égypte. In: A. Lavignac and L. de la Laurencie, eds, Histoire de la musique: Antiquité, Moyen-Age 1, Paris: Librairie Ch. Delagrave, pp. 1–34.
Lynch, T.A.C. (2018). The Seductive Voice of the Aulos in Plato’s Symposium: From the Dismissal of the Αὐλητρίς to Alcibiades’ Praise of Socrates-Αὐλητής. In: A. Baldassarre and T. Markovic, eds, Musical Culture in Words and Images, Wien: Hollitzer, pp. 715–729.
Majnero, C. (2010). The Discovery of a Probable Lingula from a Roman Tibia. In: Studien zur Musikarchäologie (Orient-Archäologie) 7 (25), pp. 39–46.
Majnero, C., and Stanco, R. (2010). Aulos – Suplu – Tibia. Uno sguardo sulle ance del mondo antico. In: M. Carrese, E. Li Castro and M. Martinelli, eds, La Musica in Etruria. Atti del Convegno Internazionale, Tarquinia 18–20 Settembre 2009, Tarquinia: Comune, pp. 121–123.
Mathiesen, T.J. (1999). Apollo’s Lyre: Greek Music and Music Theory in Antiquity and the Middle Ages. Lincoln/London: University of Nebraska Press.
Najock, D. (1996). Aristoxenos und die Auloi. In: R. Faber and B. Seidensticker, eds, Worte. Bilder. Töne, Würzburg: Königshausen & Neumann, pp. 59–76.
Paquette, D. (1984). L’instrument de musique dans la céramique de la Grèce Antique: études d’organologie. Paris: Diffusion de Boccard.
Pavolini, C. (2015). La musica e il culto di Cibele nell’occidente romano. Archeologia Classica 66, pp. 345–376.
Polychronopoulos, S., et al. (2021). Physical Modeling of the Ancient Greek Wind Musical Instrument Aulos: A Double-Reed Exciter Linked to an Acoustic Resonator. IEEE Access 9, pp. 98150–98160. DOI: 10.1109/ACCESS.2021.3095720.
Psaroudakēs, S. (2002). The Aulos of Argithea. Studien zur Musikarchäologie (Orient-Archäologie) 3 (10), pp. 335–366.
Psaroudakēs, S. (2008). The Auloi of Pydna. Studien zur Musikarchäologie (Orient-Archäologie) 6 (22), pp. 197–216.
Psaroudakēs, S. (2013). The Daphnē Aulos. GRMS 1, pp. 93–121. DOI: 10.1163/22129758- 12341239.
Psaroudakēs, S. (2014). The Aulos of Poseidōnia. In: Musica, culti e riti nell’occidente greco, Pisa/Roma: Istituti Editoriali e Poligrafici Internazionali, pp. 117–129.
Raffa, M. (2016). Claudio Tolemeo. Armonica. Con il Commentario di Porfirio. Milano: Bompiani.
Reinach, T. (1912). Tibia (Αὐλός). Dictionnaire des antiquités grecques et romaines 9, pp. 300–332.
Schlesinger, K. (1939). The Greek Aulos. London: Methuen.
Smith, H. (1904). The World’s Earliest Music: Traced to Its Beginnings in Ancient Lands. London: W. Reeves.
Sutkowska, O. (2010). Etruscan and Greek Double Pipes: An Iconographical Comparison of Their Organology. In: M. Martinelli, M. Carrese and E. Li Castro, eds, La musica in Etruria. Atti del convegno internationale, Tarquinia 18–20 Settembre 2009, Tarquinia: Comune, pp. 79–92.
Sutkowska, O. (2015). Archäologische Auloi/Tibia Funde mit Mechanismen. Die Tonkunst 9, pp. 412–422.
Terzēs, C., and Hagel, S. (2022). Two Auloi from Megara. GRMS 10.1, pp. 15–77. DOI: 10.1163/22129758-bja10040.
Wallace, R.W. (2003). An Early Fifth-Century Athenian Revolution in Aulos Music. HSPh 101, pp. 73–92.
West, M.L. (1992). Ancient Greek Music. Oxford: Clarendon Press.
Wilson, P. (1999). The Aulos in Athens. In: S. Goldhill and R. Osborne, eds, Performance Culture and Athenian Democracy, Cambridge/New York: Cambridge University Press, pp. 58–95.
Wysłucha, K. (2019). Echoes of the Rejection of the Aulos in Augustan Poetry. GRMS 7.1, pp. 105–127. DOI: 10.1163/22129758-12341336.
In spite of early studies (most extensively Bartholin 1679; on reeds, 61–3), the aulos had long remained a rather mysterious instrument even disregarding controversies concerning its reed type. Its classification as a ‘flute’, once correct when that term was quite generally applied to woodwind instruments, has spawned misconceptions up to the present day.
The first study devoted to this issue was a very brief article by Albert A. Howard (1899). See also Curtis 1914; West 1992, 83f.; Sutkowska 2010, 79–81; Hagel 2020, 428–35.
Cf. Wilson 1999; Wallace 2003; Lynch 2018.
For selective lists of aulos finds, see Psaroudakēs 2002 (Archaic, Classical and early Hellenistic period); Sutkowska 2015 (late Hellenistic and Roman Imperial period).
For a more extensive overview of various reeds in Mediterranean traditional music, see Majnero and Stanco 2010.
Smith 1904, 112: “When M. Mahillon first investigated these flutes, he supposed that the Arghool reed had been used by the players in their day; but he now tells me that, having in more recent years made the acquaintance of most of the pipes of the middle ages the cromornes, the courtauds, the dolziana, racket and others he has come to the conclusion that the Pompeian flutes were blown by some sort of double reed, but differing from the oboe and bassoon type, which are adapted on a short metallic tube of small bore; and he considers that probably they were of the sort now existing in the Japanese pipe called the Hichirichi”. A strong voice for double reeds (albeit of wrong construction) was Reinach 1912, 306f.
Cf. Naples inv. 9559, 85182 as well as the Euterpe from Moragine (cf. n. 54 below); cf. e.g. Byrne 2000; also, Loret 1893; Closson 1930.
For more details on the debate between supporters of single and double aulos reeds, see Hagel 2020, 430–5.
Cf. Baines 1967, 193.
Research on this object is underway. Some preliminary findings have been published on the webpage of the Istituto Centrale per il Restauro, http://www.icr.beniculturali.it/pagina.cfm?usz=5&uid=67&rid=50&rim=159 (accessed on 22 June 2021). Preliminary analysis suggested that the piece dates back to antiquity (Cristina Majnero, Roberto Stanco, personal communication).
Another find from the Vesuvian area has been interpreted as “lamellae” of a double reed in Majnero 2010. Unfortunately, this identification does not withstand scrutiny. Displaying green discoloration indicative of contact with bronze, the item apparently formed part of a bronze-clad wooden pipe core of which other fragments exist as well; cf. Figure 2.
Loret 1893, 98 Fig. 2; Loret 1913, 20f.; cf. Schlesinger 1939, 38 n. 2 and 404 n. 3.
Cf. Hagel 2010, 83.
For further examples, see e.g., London E164, E189, E191, E362 (Bundrick 2005, 15; 41; 77; 113). For bulbous sections, e.g. London E38, Louvre G138; New York 24.97.28; 06.1021.178; 25.10.32; Louvre G467, cf. West 1992, 85. Sources of vase images evoked in this article vary. The majority come from the Beazley pottery database, which is accessible online at https://www.beazley.ox.ac.uk/carc/pottery. Other images were spotted by the authors in museums, various publications and on the internet.
Mathiesen 1999, 186 fig. 20. Cf. Landels 1964, 394; West 1992, 85; Landels 1999, 27; 32. It must be noted that the concept of two bulbs, and especially two identical bulbs, seemed suspicious from the onset, as this design is not attested even by the iconographic examples that were evoked by Landels (1999, 27 – presumably Würzburg HA120) to prove its existence. On closer inspection, it occurs that two bulbous structures are hardly ever equal in size – the ‘upper’ bulb is usually inserted into a slightly flaring top of the (often larger) ‘lower’ bulb; cf. e.g. London E38, London E68, New York 1980.11.10, Boston 01.8082, Cabinet des Médailles 594. In the past, some attempts were made to justify the potential occurrence of two bulbs; for instance, Dietmar Najock (1996) proposed that decreasing diameters enabled curious modes of overblowing. However, being contradicted by all available archaeological evidence, none of these theories stood their ground.
In this article we use the term ‘stem’ for the cylindrical section below the constriction of a double reed. The names of this part vary depending on the reed type; in modern Western double reeds on conical instruments such as the oboe and the bassoon, the equivalent part is called ‘reed stem’ or ‘staple’. Polychronopoulos et al. 2021, designate it as “reed backbore”. The terminology of aulos sections still calls for standardisation.
Hagel 2020, 433–35. Cf. Amigues 1989, 281.
For a similar representation of bulbs, see e.g. Munich 2646 and note 37 below.
A drawing of a double-reed profile can be found in Paquette 1984, 31. The author suggests that two different kinds of aulos might have been in use – played with single and double reeds. Interestingly, one of the representations identified by Paquette as showing single reeds (p. 32) resembles in fact his sketch of a double reed profile (Tampa 1986.093, left pipe).
Smith 1904, 78–81.
Bundrick 2005, 34. The gesture of adjusting reeds seems to have been a popular motif in vase painting; cf. e.g. Oxford (Ashmolean) C207EF, London E461, private collection ABV 20340 (cf. Paquette 1984, 55 fig. A 45), Vatican 16509. It is difficult to identify the various gestures depicted in this manner with certainty – do they represent inserting reeds into or detaching them from the instrument, or perhaps putting a cap on them after performance (which appears less likely, as caps are not discernible on any of these representations)?
Becker 1966, 57f.
For more depictions of reed blades seen from the side, see belly amphora Bologna 151.
See n. 14 above.
For a similar image of the upper aulos end consisting of several bulbous sections, which is being prepared for a performance, see depictions mentioned in n. 21.
Cf. https://www.europeana.eu/de/item/09102/_RMAH_94413_NL, retrieved 2022-08-20.
For a similar depiction of a reed, cf.: kylix Yale 163; Bologna 210236. For a similar pipe design, cf. the much later ‘Wooden Pipes’ from Meroë, Hagel 2019, 183; 194 fig. 10.
Cf. Goulaki-Voutyra 2020, 366f.
A fairly similar representation of the reed is found on a kylix attributed to the Antiphon Painter (Basel BS 1961.233), showing two bulbous shapes with a triangular/trapezoidal top section.
Hagel 2020, 432; cf. West 1992, pl. 17. For another depiction very likely showing a pair of pipes linked by a ligature, see Naples 9015 (cf. Naples 111473). Reed caps are most famously shown on a mosaic in the Musei Capitolini, Rome MC0392. For earlier images, see also a calyx-krater (Louvre G103) and a very rare example on a black-figure olpe (Basel Kä 411).
Similar strings are shown in several fourth-century images, e.g. Taranto 28246; Vienna IV 1009.
Smith 1904, 74; Becker 1966, 50. Another example is Würzburg 521, attributed to the Christie Painter.
In black figure painting, the sections were sometimes signalled by strokes of white paint on a black main tube of the instrument, e.g. New York 53.11.1; London B103.16. Also, the shapes of bulbs and reeds are distinguishable in more detailed images, e.g. Basel Kä 421.
A bulb section of a similar shape was excavated as part of the so-called Lindos aulos, dated to 525 BC; cf. Blinkenberg 1931, pl. 16 no. 450; Bélis 1988, 239.
For a chronological overview of aulos finds, see Psaroudakēs 2002, 349 and 356.
For images of these items, see Psaroudakēs 2013, 109f.
Cf. also e.g. New York 1993.11.5; Basel 491; Boston 1973.88.
Reed stems in black paint in contrast to a white body of the instrument are also found on a black-figure amphora attributed to the Amasis Painter (Basel Kä 420).
E.g. another kylix by Douris (Florence 3922), an oinochoe attributed to the Methyse Painter (Louvre G440), a kylix attributed to the Antiphon Painter (Munich 2635), a stamnos attributed to the Menelaos Painter (New York 06.1021.178), a skyphos attributed to the Brygos Painter (New York 1980.11.10); Louvre G467; National Museum, Acropolis Collection Athens 2.5.
For the Poseidonia aulos bulbs, see Psaroudakēs 2014; Bellia, Pavone 2021, 389. Similar slender ‘elongated’ bulbs seem to have been part of the aulos of Ialyssos. Unfortunately, the dating of this find is imprecise; thus it cannot be confidently placed within the fifth-century, cf. Psaroudakēs 2013, pl. V 3f.
Cane auloi are attested e.g. in Theophr. Hist. plant. 4.11.3, where the term bombykías must refer to cane usable for the instrument’s body (bómbyx), as opposed to zeugítēs, “mouthpiece cane”.
E.g. New York 25.78.66, Würzburg HA120.
Cf. Psaroudakēs 2013 (on the Elgin pipes, 101 n. 31). The wooden parts from the Giglio shipwreck lack an upper section, which appears to indicate that separate upper parts would have been added. These might reasonably have reflected the neck-within-socket design we observe in bone instruments.
Cf. Psaroudakēs 2008; Banou 1998. For other reed sockets detached from the bulb, see one aulos from Megara and the aulos fragments in Lecce, cf. Majnero and Stanco 2010, 124.
Cf. Würzburg HA120; Würzburg 4781; Taranto 8263.
Cf. Psaroudakēs 2008, 209; Bellia, Pavone 2021, 384.
Terzēs and Hagel 2022, 64.
Hagel 2012a.
Cf. e.g. Hagel 2021.
Hagel 2020, 424–7.
Stelios Psaroudakēs, personal communication.
Cf. n. 30.
Especially images of pipes found on small Gnathian pottery (chous and oinochoe) in the National Archaeological Museum of Taranto and the Sigismondo Castromediano Museum in Lecce.
For an image, see Wysłucha 2019, 117.
Images such as Naples 9559 are often misleading in this respect.
Cf. Cervelli 1994, 38.
Particularly bulging reeds are also depicted in a sacrificial scene on the Parthian monument from second-century AD Ephesus (now Ephesus Museum, Vienna, Antikensammlung, I 865).
For respective images, see Bélis 1986, 23; Pavolini 2015, 360; 362.
E.g. mosaics in the Zeugma Mosaic Museum, Gaziantep (Clio and Euterpe); the National Museum of Archaeology, Lisbon (inv. 999.149.1); the Western Archaeological Area, Kos (part of the mosaic depicting the Judgement of Paris); sarcophagi in the National Museum of Rome: Palazzo Massimo alle Terme (‘Sarcofago Mattei’, inv. 252), Rome; the Ostiense Museum of Ostia Antica (Sarcofago delle Muse, inv. 59954–5); the Nelson Atkins Museum of Art, Kansas City (Muse Sarcophagus, inv. 87–21 and 33–38).
Now held at the Museo Arqueológico Nacional in Madrid, inv. 3619. A similar scene is depicted on a mosaic from Trier, Rheinischen Landesmuseum Trier, inv. 10703–10724. A respective myth is not known from any written sources.
Hagel 2008, 58; Hagel 2020, 424, respectively.
Barker 1984, 188 n. 10; West 1992, 84; Hagel 2020, 418–32 against Becker 1966, 51–80.
Cf. Chailley 1979, 62.
See above with n. 30.
The passage has sometimes been understood as referring to some kind of oil (Barker 1989, 105) or grease instead of saliva (adopted in Barker 2015, 237; Raffa 2016, 768f. n. 408), apparently based on a misreading of the relevant lexical entries in LSJ. Whether one reads neuter
For this device cf. Hagel 2012b; for a newly documented mechanism from the period of the text, Terzēs and Hagel 2022. The present passage appears to interpret its effect as subtracting a part of the air stream, which escapes through the small side hole, so that the rest is no longer “in abundance and dense” and does not really “fill the pipes”.
A virtually certain correction (a neuter plural is required) for the manuscripts’ meaningless
Note that the old idea of a mouthpiece inserted sideways (already Howard 1893, 16; cf. Düring 1934, 172) is actually incompatible with the text, where it is only the blades that are placed at the side of something described by a neuter noun.
Modern Western experience may not be a reliable guide, though, because our double reeds are typically coupled with a conical resonator and therefore play an octave higher as well. On the other hand, exceptionally massive double reeds, such as that of the duduk, may also produce a ‘soft’ sound, especially when attached to a cylindrical resonator.
Hagel 2018.
For the passage in general, see Barker 1984, 186–9.
The term is an emendation of the manuscripts’
For the meaning of
Cf. West 1992, 367–9.
Antyll. ap. Oribasius Coll. med. 6.8.5–6:
Sext. Emp. 6.41–42.
Byrne 2002. Note that the author, even though discussing plásis, does not bolster his argument by reference to kataspásmata. It may be worth noting that a related explanation of kataspásmata as pulling down metal sliders that closed holes in the bass region and thus producing pitch jumps cannot presently be reconciled with the evidence. While -spân for operating a slider is well attested (Aristox. Harm. 1.21, 27.1 Da Rios, [Arist.] Aud. 804a), the known aulos models from the period in question do not support a melodic use of their sliders, because a higher fingerhole will always be opened while switching a slider state; cf. Terzēs and Hagel 2022, 33–5. Therefore, the action of pulling/drawing a slider and pitch change seem to have been dissociated in Theophrastus’ period.
Note however the contrast to the theorists’ insistence on a stable pitch as the defining characteristic of a note (phthóngos) from Aristoxenus on (e.g. Aristox. Harm. 1.15, 20.15–19 Da Rios). But Aristoxenus himself otherwise reports severe clashes between theory he upholds and melodic practice, as in the notorious case of microtonal adjustments of notes whose pitch should have been ‘fixed’ (ap. [Plut.] Mus. 1145d).
In Quintilian cf. e.g. 3.8.26 mansuetum quoque (sic enim sunt interpretati
Cf. also Pers. 1.17, where plasma is associated with the throat and called ‘liquid’.
