A Welsh term for a plucked and, from about the 11th century, a bowed lyre. The name is cognate with the Irish crot, cruit, which originally denoted a plucked lyre but was ultimately used for a harp. The Middle English crouthe, crowd(e) is a late 12th-century borrowing of the Welsh crwth. For related north European lyres see Rotte (ii), Stråkharpa, and Scotland, §II, 8.
Three 18th-century six-string Welsh crwths have survived: the Foelas crwth (fig.1), made in 1742 by Richard Evans (fl 1736–56); one held at the Warrington Museum and Art Gallery; and a third at the National Library of Wales, Aberystwyth (fig.2). A small number of modern makers now make exact copies of crwths, and playing techniques are being rediscovered through experimentation and in the light of the evidence of 16th-century treatises.
2. Technical and theoretical considerations.
BETHAN MILES (1), ROBERT EVANS (2)
Although the three 18th-century instruments vary slightly in proportion and design they share the same basic structure, the body, including the neck, being carved from a single block of sycamore in the age-old manner with the sound-board made of pine. To lighten the upper part of the instrument the insides of the arms were hollowed out, echoing Dark Ages lyre construction. The string holder at the back of the yoke was also hollowed out. The forward sloping arms are found in numerous depictions of classical lyres, and on the crwth have the effect of keeping the strings parallel to the sound-board. The Foelas and Warrington crwths share some refinements not seen on the Aberstwyth instrument, which is slightly smaller in size. A narrow baulk of sycamore, part of the body of the instrument, is visible from the front and flush with the surface of the sound-board. The sound-box is wider than the frame (see fig.1), which would not afford the extra support mentioned below (§2) were it widened to continue flush with the sound-box. Both instruments have mock purfling along the front outline. None of the crwths has a brass bar and the blocks in the Foelas crwth are probably modern. The Warrington crwth has more slender arms than the other twos. This carefully crafted crwth with its decoratively shaped yoke and, unlike the other two, a separate nut placed on the yoke to protect the lateral drones, has neither a saddle nor sign of wear from tailgut. Unfortunately the tailpiece, the fingerboard and nut, and the bridge are missing; in fact the other crwths are exhibited with modern curved bridges. The heads of the tuning-pins on the Foelas crwth are filed with decorative crosses and some features - such as the neck button and fingerboard - were influenced by violin construction that perhaps reflects Richard Evans' experience as a violin maker. Valuable information regarding the crwth in 16th-century Wales is provided by a panel of a 16th century bedhead at Cotehele House, Cornwall, showing a Welsh crythor and harper, and by references in Welsh poetry, especially the strict-metre request poems in which an object was described by means of numerous comparisons. Lack of pre-16th-century information in Welsh sources makes it impossible to confirm whether the crwth had always been six stringed or had at least become so by the time the apparently six-string crwth was carved on a misericord at Worcester Cathedral in about 1397. Unlike the harp and the medieval fiddle, the crwth did not develop beyond its late medieval form.
The crwth was a rare instrument in medieval England, apparently already obsolete by the 16th century, although it retained its high status in Wales until the end of the century. Even its popularity in 14th-century England may have been due to its favoured status at the courts of Edward I and II: over half the crwth players (croutheres or crouderes) mentioned in the expense accounts were from Wales and the Welsh Marches. Significantly, three depictions of the instrument played by or associated with minstrels date from this period: Book of Hours from York c1300 formerly in the possession of C.W. Dyson Perrins (MS 12.f.76), the seal of Roger Wade, 1316 (GB-Lbl seal no.lxxxvii.44), and a treatise on Kingship by Walter de Milemete (GB-Och 92.f.43). With the exception of the original 15th-century carving of a crwth (incorrectly restored after storm damage in 1894) at St Mary's Church, Shrewsbury, which has a figure-of-eight form (see Galpin, 1910), these and other depictions resemble in shape, if not in detail, the later Welsh crwth, having a straight-sided form with one or both ends rounded.
The paucity of pre-16th-century iconographic sources for bowed crwths necessitates a particularly cautious approach in evaluating the evidence of the number of strings and lateral drones in view of the difficulty of accurate depiction, especially on carvings (for a discussion of these problems, see Fiddle). The problem is compounded by ambiguities in terminology in regard to the medieval crwth (see Rotte, (ii), Chorus, (iii)). Exactly when the bow began to be used on north European lyres is not known. Bachmann hypothesized that the spade-like fiddle common in southern Europe in the Romanesque period evolved from the lyre, which in that region had developed a neck in the 8th–9th century. Whether north European lyres already had a neck before the bow was adopted is open to speculation. Three-, four- and five-string waisted oval lyres played with bows are depicted in 11th- and 12th-century continental sources, some having necks, but not always with centrally set strings.
The lyres depicted in the Winchcombe Psalter (c1030–50, GB-Cu Ff.1.23, f.4v) and in an early 12th-century Durham manuscript (GB-DRc Hunter 100, f.62v) raise difficult questions of interpretation. The Winchcombe Psalter shows a plucked three-string lyre as well as a bowed four-string lyre held at the shoulder; in both cases the strings would appear to be stopped over the end of the instrument rather than through an opening, although it may be that the instrument is merely being held, rather than fingered (fig.3). However, the four centrally set strings of the bowed instrument would imply the presence of a neck. It may be that the artist wished to show his knowledge of a new feature, i.e. the bow, which is quite carefully drawn. Although a bow is not included with the Durham lyre, it has four centrally set strings, which would suggest a neck, and a further two set at an oblique angle (fig.4). This six-string lyre tantalizingly resembles the 18th-century crwth of fig.1 in so far as it would appear to have a similar string formation, albeit here depicted in mirror image. However, in view of the fact that this instrument clearly illustrates the word ‘lira’ in a treatise on the constellations, it may be that the artist had seen a four-string bowed lyre, but thinking that a lyre ought to have six strings (see Rotte, (ii)), added two more, thereby conforming to Pythagorean symbolism. A carved stone figure of c1200 (now much weathered) at St Finian's Church, Waterville, Co. Kerry, plays a rectangular apparently six-string bowed lyre.
As with the medieval fiddle, the few pre-16th-century depictions of the crwth have variant features regarding the bow, the methods of attaching the strings at the lower end, the presence of a bridge and fingerboard, and the number of strings. In the stained glass windows (1447) at the Beauchamp Chapel, St Mary's Church, Warwick, two five-string crwth players are seen facing a pair of harpers and therefore at a slight angle, enabling us to see that the neck is thin. Here, as in the 15th-century carving shown in fig.5, the crwth players' fingers press directly on to the neck. The absence of a fingerboard meant that a separate bridge was unnecessary and that a frontal string-holder or (as seen in fig.5) a tailpiece resting directly on a bridge or possibly tilted upwards with a wedge, would be sufficient. As shown by 18th-century evidence (see Barrington, also fig.1), the two most reliable crwth depictions, the seal of Roger Wade (1316) and the Welsh carving on a 16th-century bedhead now at Cotehele House, Cornwall, clearly show a flat bridge, which usually implies flat fingerboard – features that are difficult to depict in relief. The 18th-century bridge was placed obliquely with one foot extending through a right soundhole and standing on the back, thus acting as a soundpost. While the bridge foot enters the opposite circular soundhole on the Cotehele crwth, this feature is not obviously visible on the seal of Roger Wade. Some south-east European folk fiddles today have a combined bridge and soundpost (see Bridge, (i)).
That the crwth of the official 16th-century Welsh crythor was a six-string instrument is confirmed by the request poems and the surviving 18th-century crwths. An anonymous poem addressed to Robert Rheinalt, a master crwth player, who appears in the court records of Henry VIII as a ‘Welsh minstrel’, requesting a crwth on behalf of Edward Grythor of Yale, gives a description of the string arrangement which corresponds with the 18th-century evidence of a six-string crwth tuned in three pairs of octaves:
Ei
ffrismal a ddfalwn
A thri sydd I wneuthur sŵn:
Crasdant, cywirdant fal cynt
A'u bwrdwnau'n ber ‘dantynt;
…
Lle i'r fawd yw'r llorf a'i was;
(We shall describe its principal [strings], and there are three which produce sound: the upper and middle as before, and their bourdons sweetly sounding below … the lower string and its servant are the thumb's realm)
Although the poem does not mention the specific tuning of G C D, by examining the cerdd dant (string craft) treatises alongside the Robert ap Huw manuscript one can see that cywirdant/cyweirdant would correspond with C, crasdant (D) and llorfdant (G) (see §2 below). However, apart from the six-string Worcester misericord crwth (c1397) and the above-mentioned Durham manuscript, other medieval and Renaissance depictions would appear to have three, four or five strings. Because of its Welsh connections, the Cotehele crwth might have been expected to have six rather than five strings. While the craftsman perhaps had difficulty in carving six strings in relief in the space available, one must seriously consider that at least one of the Beauchamp Chapel crwths has its five strings arranged in one single and two double courses, all of which appear to run above the neck. What the small number of crwth depictions reveals, however, is that over half, including the Roger Wade seal, are four-string crwths.
The use of ‘crwth’ as a generic term for bowed instruments from the late 16th century onwards can cause difficulties in interpretation. As in English, the Welsh poets sometimes used ‘fiddle’ or ‘fiddler’ as terms of abuse, as in a light satirical poem in which ‘crwth’ and ‘ffidl’ are used interchangeably. A reference to a ‘fiddler a chrwth trithant’ (‘a fiddler who plays a three-string crwth’) (GB-Ab Peniarth 77.p.175) being one of four types of inferior entertainer should not necessarily be taken as direct evidence of a three-string crwth. Moreover a sketch of a three-string crwth and bow, coupled with harp and two bagpipes, used to illustrate a list of the various ‘ancient’ grades of bardic craft in an early 17th-century manuscript (GB-CDp 2.634, f.358) was perhaps inspired by humanist ideas. It may reflect a 16th-century Welsh version of the legend concerning the origin of the lyre in which Mercury is credited with the discovery of the crwth with three horsehair strings and a bow. However one cannot dismiss the possibility of a three-string crwth being used in Wales concurrently or maybe before the development of a crwth tuned in three pairs of strings. This may be what the poet implies by ‘fal cynt’ (‘as before’) in the above quoted poem; the six-string crwth would call for a more advanced technique (see §2 below). The three-string crwth of a 1498 carving at Milton Abbas, Dorset (fig.5) seems to have a lateral drone touched by the thumb.
With the possible exception of the Dorset carving (fig.5), lateral drones are not obviously depicted in any of the sources. On the Roger Wade seal the lower pre 18th-century string seems to be thicker and set slightly apart from the other three; on the Cotehele crwth the fourth string runs at a more oblique angle. On the Worcester crwth, which is held in the same downward-pointing position as the Beauchamp Chapel crwths, the thumb seems to be pressing down on the lower two strings. Conversely it may be that these particular crwths are merely being held, not played. The poem addressed to Robert Rheinallt also refers to ‘seven tuning-pins, one of which is unused/spare’: ‘Llwyn tew o ebillion teg; Os aeth enw, saith o honym’, Eisiau gradd, segyr yw un’. This tallies with another poem requesting seven pins: long, straight metal tuning-pins which fitted tightly were obviously much valued.
Although there is no surviving iconographical evidence of the crwth in Wales from before the 16th century, there is a considerable body of literary and documentary evidence which attests to the important role the crwth played in Welsh society as a high-art instrument. The earliest reference to ‘crwth’ or ‘crythor’ is found in a pre-1100 poem:
Wyf
bard ac wyf telynawr
Wyf pibyd ac wyf crythawr
(I am a bard, a harper, a piper and a crwth player).
The same three instruments are referred to in Brut y Tywysogion (‘The Chronicles of the Prince’) at a special feast held at the court of Lord Rhys in 1176 and also by Giraldus Cambrensis. These references to a trinity of musical instruments, echoed in the early 17th-century manuscript mentioned above, might be interpreted as a mere formulaic literary device. If one could prove the hypothesis of a formulaic triad of plucked, bowed and blown instruments, this poem would take the earliest evidence of a bowed crwth in Wales back to the 11th century. Nevertheless, the references show that the crwth was well established at an early date, although its stringing is unknown.
The crwth and harp were the only two instruments used in Welsh bardic performance throughout the later Middle Ages, although the greater number of elegies and request poems involving harp/harper reflect its higher status. The 16th-century Statud Gruffud ap Cyanan (‘The Statute of Gruffudd ap Cynan’) (GB-Lbl Add.19711, second quarter of the 16th century), written mainly to emphasize the superiority of the guild of poets and musicians at a time when the authorities were legislating against unlicensed entertainers, contains valuable information regarding the requirements of the various grades of bardic craft. Harpers and crwth players largely shared the same repertory. However, it may be significant that, in one of the 16th-century manuscripts listing titles of musical repertory (GB-Ab 1711b.f.62r, 66v), separate named musical pieces are given for the cadair and the colofn, two of the most technically advanced genres. Some later 16th-century versions of the Statud state that the datgeiniad (‘declaimer’)
Towards the end of the 16th century the professional poets' stubborn reluctance to forgo the old strict metres in the face of competition from free verse based on the metres of popular English airs meant that they eventually lost the patronage of the increasingly anglicized Welsh nobility, who embraced the type of musical entertainment and instruments fashionable at the Elizabethan court. There is no doubt that many of the professional musicians, in order to retain their status, adapted themselves to playing the currently popular plucked and bowed instruments. While the crwth continued to be played by a lower stratum of musicians patronized by a correspondingly lower social stratum, the inability of the crwth to evolve and compete with the increasingly popular violin caused its eventual demise towards the end of the 18th century, by which time it had become an object of antiquarian curiosity. Daines Barrington's sketch of John Morgan, one of the last crwth players in north Wales with the caption ‘Method of holding and playing on the instrument’ (c1770; Society of Antiquaries, London), showing the method of supporting the crwth with a strap, corresponds precisely with evidence from 16th-century poetry and iconography and possibly with the minstrel depicted in a Book of Hours from York of about 1300 (formerly in the possession of C.W. Dyson Perrins).
Barrington in 1770 recorded the crwth's tuning as g–g'–c''–c'–d'–d'', but when Bingley gave the tuning as a–a'–e'–e''–b'–b'' over 30 years later, the player he heard had obviously adopted the fiddle tuning in 5ths. By exchanging the flat fingerboard and bridge for curved ones, as clearly happened with the Aberystwyth crwth, it effectively became a ‘violon en forme de crouthe’.
Experiments using a six-string Welsh crwth played slung from the neck reveal many of its anomalous features to be highly practical. The arched shapes cut from the yoke on either side of the fingerboard allow space for the fingers to reach the semitone next to the nut (half position in modern terms), and for the thumb to pluck the lateral strings. The oblique angle of the bridge to the nut aligns with the angle made by the fingertips against the fingerboard, making it possible to play in tune using the basic technique of the crwth: stopping each octave pair of strings with one finger. The stopped strings, though tuned in octave pairs (c''–c'–d'–d''), are not grouped into courses but evenly distributed across the width of the fingerboard, giving space for each string to be stopped separately if required, thereby splitting the pairs to make richer chords or allowing one of the pair to complete a harmony while figuration is played on another.
The two strings (g–g') lying off the fingerboard are on a slightly different plane from the stopped strings, which takes them out of the bow's reach except when playing close to the bridge, which is flat (the curved bridge shown on the Foelas crwth in fig.1 is not original). This allows three combinations: sounding all six strings at once with the bow close to the bridge; bowing the four stopped strings and leaving the remaining two silent; or bowing the stopped strings while plucking the two G strings with the thumb. The last feat is made easier by opening the left hand to bear against the surrounding frame, thus giving an extra point of contact and stability.
Until the end of the 16th century the crwth's special status alongside the harp meant that a prestigious, formal repertory analogous to that of the harp (see Harp, §V, 1(i); and Robert ap Huw) was also played on the crwth. The harp tablature in the Robert ap Huw manuscript (GB-Lbl Add.14905) contains pieces composed on the 24 measures of string music (pedwar mesur ar hugain cerdd dant). These are successions of two contrasting sets of notes – principal and weak – which predominate alternately, forming binary patterns represented by the figures ‘I’ and ‘O’, e.g. corffiniwr IIOO IOII. By convention, crwth players represented the measures differently, reversing the figures. Corffiniwr thus became OOII OIOO, distinguishing between different branches of the craft guild and reflecting the fact that a crwth player is able to play the notes c'–a' (Guido of Arezzo's ‘natural Hexachord’) in the first position but is unable to complete the scale downwards to g, as is a harper. A harper, therefore, is able to use the most principal of Guido's hexachords, which climbs from T and provides the first triad of the gamut: T, B, D; occupying the lowest three lines of the ‘great stave’ (‘I’). The first complete hexachord available to a crwth player climbs from C, it is less principal, and gives the triad: C, E, G; occupying the second, third and fourth spaces from the bottom of the ‘great stave’ (‘O’). In this context the lines are equated with ‘I’ and the spaces with ‘O’ (Peniarth 155) rather than corresponding to fixed or movable pitches respectively as outlined below.
A crwth player interprets the harp music in the Robert ap Huw manuscript by identifying the predominating set of notes (i.e. the principal set or those marked ‘I’ in harp terms) by playing the open C strings while playing figuration on the D strings. When the weak set (those marked ‘O’) predominate, the arrangement is reversed. The plucked G strings generally form a bass common to both sets.
The open strings of the crwth correspond to the cyweirdannau (principal or fixed strings) of the harp. The notes e', f', a' and b' and their octaves, produced by stopping the crwth's strings, are the lleddfdannau (weak or movable strings) of the harp. The crwth's tuning is well suited to realizing the binary patterns found in the Robert ap Huw manuscript, and fulfils the Pythagorean ratio 12:9:8:6, as do the fixed strings of the harp, recalling the hestotes (fixed elements forming a 4th, a 5th and an octave) and kinumenoi (movable elements: the interior notes of a tetrachord) of ancient Greek music. Tuning in octave pairs enriches the sound and is a built-in arrangement for converting 4ths into 5ths (5ths were considered more perfect consonances than 4ths during the Middle Ages). This tuning also produces ‘octave ambiguity’ and most closely resembles the third of the three tunings for the viella given by Hieronymus de Moravia (see Page).
The most important treatise dealing with bardic music for harp and crwth, Llyfr Cadwedigaeth a Dosbarth Cerdd Dannau (see Peniarth MS 155, also Miles), contains directions to the crwth player for creating the five standard and warranted cyweiriau (harp tunings; see Robert ap Huw) by stopping the strings. This source and others give directions for tuning the harp in the five standard and warranted tunings in terms of the Pythagorean tradition and the Guidonian system. Robert ap Huw gives tuning-charts for two of them (cras gywair and lleddf gywair y gwyddil; see Lewis, 108–9). By applying the index and middle fingers appropriately to the crwth's fingerboard, the positions of the semitone (mi–fa) are identified within the C–F tetrachord. F (fa) is omitted in the two pentatonic tunings (see Lewis, 7).
In later medieval Wales, the most splendid production and performance of bardic poetry and music involved four professionals: a poet, a harpist and or a crwth player, and a datgeiniad (declaimer). Two or three of these roles were sometimes played by one person, especially near the end of the 16th century, when patronage for bardic poetry and music became meagre.
During the 17th century crwth players had to play the new, pan-European, bourgeois music demanded by all classes. They played jigs and other dance tunes, and accompanied popular songs, ballads and post-Reformation carols. Remnants of bardic repertory survived briefly outside their social context. When it died out at the end of the 18th century the crwth still embodied the musical theories of ancient Greece and medieval Christianity and of Welsh bardic music in its lyre construction and characteristic tuning. Modern players using accurate copies of crwths and using bridges which are, in Barrington's words, ‘perfectly flat, so all the strings are necessarily struck at the same time, and afford a perpetual succession of chords’, have begun exploring the bardic repertory, accompanying traditional Welsh and English popular songs and ballads and playing historical dance music.
GB-AB Peniarth 155 (16th–17th century), 79–83
D. Barrington: ‘Some Account of Two Musical Instruments used in Wales’, Archaeologia, iii (1775), 30–34
W. Bingley: A Tour Round Wales during the Summer of 1798 (London, 1800)
F.W. Galpin: Old English Instruments of Music (London, 1910; rev. 4/1965 by T. Dart)
H. Lewis, ed: Musica (B.M. Addl MS.14905) (Cardiff, 1936) [facs. of the Robert ap Huw manuscript]
A.O.H. Jarman: ‘Telyn a chrwth’ [The harp and the crwth], Llên Cymru, vi (1960–61), 154–75
W. Bachmann: The Origins of Bowing (London, 1969)
J.M. Bevil: The Welsh Crwth: its History, and its Genealogy (diss., North Texas State U., 1973; suppl., Houston, 1979)
C. Page: ‘Jerome of Moravia on the Rubeba and Viella’, GSJ, xxiii (1979), 77–98
B.E. Miles: Swyddogaeth a Chelfyddyd y Crythor [The function and art of the crwth player] (diss., U. of Wales, 1983)
M. Remnant: English Bowed Instruments from Anglo-Saxon to Tudor Times (Oxford, 1986)
R. Evans: ‘The Crwth’, Early Music Today, vi (1998), 18–19
R. Evans: ‘Experiments with the Crwth’, GSJ (forthcoming)