The Spanish Guitar in the Nineteenth and Twentieth Centuries Author(s): Terence Usher Source: The Galpin Society

Journal, Vol. 9 (Jun., 1956), pp. 5-36 Published by: Galpin Society Stable URL: http://www.jstor.org/stable/841787 Accessed: 18/07/2009 11:16
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=gal. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org.

Galpin Society is collaborating with JSTOR to digitize, preserve and extend access to The Galpin Society Journal.

http://www.jstor.org

TERENCE

USHER

The

Spanish and

Guitar

in

the

Nineteenth

Twentieth

Centuries
to of but purport be a history the instrument, dealsonly with the of morerecentaspects its designandconstruction sincethe end of the That particular hasbeen chosenbecause of century. eighteenth point three significant modifications which took place in a few decades aroundthattime-the additionof a sixthstring,the increase body of andthe adoption whathasalwaysbeenknown as the 'Spanish of size, construction. style'of internal The linesof improvement the Spanish of guitarhavefollowedthe of developments the stringed of in instruments general: aim the pattern hasbeento increase sonority carrying the and of theinstrumentso power thatthe guitar its in to maymaintain comparative position regard tone andvolume,withinthe general of increase soundlevelsin music,asin the caseof the transition the bowedinstruments theviol family in from to theviolinandcello,andin thekeyboard instruments thewoodfrom framepianoforte the iron-frame to pianoforte. The outcomeis thatthe guitarof the lateeighteenth earlynineand teenthcenturies-an instrument modestcapacityhardlysuitedto of public performance, though frequentlyso used by Sor, Carcassi, Carulli,Aguado, Coste and other virtuosi-is becominga concert instrument considerable of volume and power in competenthands, with somelimitations) guitar of concerti with small capable present (at and aural orchestras, of holdingits own on termsof reasonable equality with the chamber groupsof today. The luthier-designer at all timesconfronted is with the fundamental if to problemthatit is difficult, not entirely impossible, makea guitar whichis at one andthe sametimeidealforplayingin the smallmoder homeandon the concert of but platform a largehall.The subtle, clear, characteristic of the guitarwhichhasa comparatively tone simplefanto wood strutting systemconfined thelowerboutandfairlysubstantial in thickness the belly,is idealfor home playing;but uponthe concert of platform,this clear,ringingtone deteriorates, absorption the by lowerpartials, a tinnytinkleof sound.All elsebeingequal, larger to the the enclosed the spacein which the guitaris played,the greater pros

of THIS outlineof the morerecentdevelopment the guitardoesnot

portionate absorptionof these lower partials,and as economic factors nowadays force the professionalartisteto play in halls holding from, at least, six or seven hundred people upwards to halls of the size of the Royal FestivalHall and the Free Trade Hall at Manchester,each holding audiences of three thousand, the luthier-designer is required to supply to the professionalartistea guitar with such a plenitudeof lower partialsthatwhen some of these have been absorbedby the hall, the resulting tone will still be deep, resonant and well-balanced. Unhappily, no luthier has solved the dilemma, and few have even effected a satisfactorycompromise: faced with the enforced choice of one or other type of tone, the luthier making a concert guitar wisely chooses to overload the guitar with lower partialsby the more complex strutting systemsand/or thinnerbelly wood later to be described,so that on the concert platform, where the artiste'sliving is made, the result will be satisfactory.But when played in a small room, the guitar'stone is then too thick for the interpretationof many works, particularlythose originally intended for the lute or vihuela. Whether any increase of volume and tonal depth in the Spanish guitarwas, in fact, desirable,must be left to the plane of aesthetics,since development is not always artisticallysynonymous with improvement. Changes in musical instrument design arise from the demands of virtuoso performers on the one had, and from the requirementsof revolutionarycomposers on the other hand. In Beethoven's increasing deafness may, perhaps, be seen the seeds of the powerful iron frame pianoforte of today. But though the works of Franz Liszt, by their breadth, and Bela Bartok, by their percussive demands, cannot be imagined as being adequately interpretedupon a Broadwood woodframe grand of I885, those of us who have heard one of these old pianofortes properly restored, or have played a moder reproduction of the Viennese fortepiano, may be forgiven if we are led to wonder whether the earliermusic can in turn be adequatelyinterpretedon the massive iron-frame concert grand pianoforte. The guitarist is today in a similar quandary:a good part of his repertoireconsists of transcriptionsof ElizabethanEnglish and Continentallute music, and many of us have graved6oubtswhether the moder guitarrade concierto can the dry, bleachedcleanlinessof tone essentialto the successful reproduce interpretationof such music. But we must leave these questions of taste to other pens. Perhapsthe truth is that a guitar can no more be a lute than a clarinetcan be an oboe, or a piano a harpsichord; if we and want to use each tone colour in particularworks, we must use both instruments. Since changesin the Spanishguitar design have not been confined to
6

it after a any particular aspectof its construction, will be necessary, survey,to dealwith eachaspectin turnin orderto appreciate general whichhastakenplace. the overalldevelopment
GENERAL CHARACTERISTICS (fig. I) By the end of the eighteenth century the guitarwas a familiarinstrument throughout Europe: guitars are still extant of that period from Spain, France, Italy, Austria and Germany, and though one has not seen examples from Scandinavia,doubtless they exist. One must be wary of generalizations,or of drawing inferencesfrom the examination even of all the limited number of guitars remaining from that period, much less those which one has been fortunateto see. But it would be safe to say that at the end of the eighteenth century, and the beginning of the nineteenth, the design of the guitar, not inappropriately, was further advanced in Spain than in any other country. Already the Spanishluthiers had discovered the importance of belly-wood as thin as that of the lute (were not the Eli7abethan lutanistsadvised periodically to remove the belly of their lutes and to iron it flat again afterthe warping of its extremely thin wood underthe string tension?) and a rudimentary system of lateral barring had replaced the transversebarring below the belly. (This question is more exhaustively dealt with under a later head.) But in other countries, though the wood and workmanship were good, luthiers retainedthe relatively thick belly and tranversebarringof earlierdays. In all countries,the guitarwas still narrow acrossthe upperand lower bouts, and comparatively wide in the waist; the many guitars in the paintings of Watteau all have this characteristic,though one realizes that some will be chitarre battente. decadeby decade,the nineteenthBut the centuryluthiersincreasedthe width of the bouts, particularly lower bout, in proportion to the body length. In Spainand Italy the length of body remained fairly constant, and the measurementsof upper bout, waist and lower bout progressively increased.But in Franceand Germany the body length in some instancesbecame a little less, the waist remained the same, and the upper and lower bouts increased,so that the general form of the guitar tended to be that of a double-circle or figure-of-eight in North Europe and of two joined ovals in South Europe. The Spanishhead, too remained vigorous and masculine, widening aggressively to the upper end, whereas the Italian and French heads, which had always widened less sharply, retainedtheir feminine characteristics.The Lacote peg head of the early nineteenth century echoes, in miniatureand in reverse,the figure eight of the body. On the whole,
7

the Spanishluthiers seem to have retained the peg head longer than Northern European luthiers, possibly because machinery to make the tuning machines was more readily accessible in North Europe where the IndustrialRevolution was more advanced. Since this article was begun, the writer, quite fortuitously, has

I|^I

S-^l

Fig. i. left Guitarby Antonius Stradivarius(I680) rightGuitar by Hermann Hauser (I929) 8

acquiredor borrowed in Manchesterfour guitarswhich enable a comparisonto be made of instrumentsfrom I790 to I850. Photographsof these instruments,together with those of modem concert guitars by CliffordEssex and Harold Petersen,appearon PlatesI and II and a brief descriptionof the instrumentsis given below:
FRENCH GUITAR, UNLABELLED, PRESUMED circa I790, known as the Cusworth Hall Guitar. Head and fingerboard faced rosewood. Ebony bridge. Satinwood back and sides. Two transversestruts, one above and the other below the soundhole. Sides and back beautifully banded in darker lacquer beneath the varnish. Table and soundhole edges inlaid with black and yellow wood, but stringing is painted on.

Dimensions: Body length Scale length

I74"

241'

8!' Upper Bout Waist 6"' Lower Bout Io0" 3" Depth at deepest point Fingerboardwidth at nut I^f Body joins neck at I ith fret Total number of frets 17 (II ivory on neck; 6 ebony on table)

Back and table of guitar are both absolutely flat. Fingerboard is slightly ovalled. The tone is powerful, but coarseand crude. This guitar, in shape and ornamentation,correspondsclosely with instrumentsin the Watteau and other paintings and engravings of the eighteenth century; the style remainedastonishinglyconstantuntil the century had turned. It may be seen from these engravings that the instrumentshad five pegs down each side of the head for the tuning of the five double strings usual at the time (see Note at end of article). The Cusworth Hall Guitar, however, has three pegs down each side of the head, but this retainsin some degree the elongatedshapedesigned to accommodate five pegs down each side. Either this indicates that the instrument was made soon after the change to six single strings, or the neck and head have been changed.But the bridge, which appears to be the original, bears only six single holes for the strings. The neck is screwed to the end block from the inside by a metal woodscrew, which may indicate a neck replacement.The general condition of the Cusworth Hall guitar is nothing short of remarkable;indeed, one does
9

not experience that indefinable 'feel' of antiquity which is apparent with other instrumentsof the time, or with the most modest Sheraton, Hepplewhite and even Regency furniture. One wonders whether this may, in fact, be a mid-Victorian reproduction, but what its purpose? Amateur theatricals Cusworth in the Jane Austen manner (Mansfield at Park) On the other hand, the guitar was bought at Cusworth Hall (near Doncaster) by a Galpin Society member, a couple of years ago. The mansion was built in I750. Amongst the effects sold at the same time as the guitar was a guitar tuition manualbearing the tide: A New Tutorforthe SpanishGuitar,containing progressive a selection of the mostFavouriteWaltzes, Rondos,Songs, Glees, etc.-composedby various authors arranged FeliceChabran. London. and Printedby .... by Potter& Co., 20 Soho Square,& to be had at Goulding,D'Almaine, Street,Dublin. 7 Westmorland The following entry: 'Chabran,a composer of vocal and instrumental music, arrangedfor the guitar, and published in London about the year I789' appearsin A Dictionaryof Musicians from the earliestto the the contents theworks time,comprising mostImportant present biographical of . with Upwards a hundred of Gerber . . together of originalmemoirs the of mostEminentLiving Musicians a Summary the History of Music. and of Vols. I and II SecondEdition.London... .1827. There is no entry in Grove (5th edition) for Felice Chabran, but under Francesco Chabran Chabran(Ciabrarro) -- d. b. appearsthe following: 'Francesco guitar player and composer. He published a Tutor for the Guitar in etc. If this tutor were bought for use with the Cusworth I790....' Hall guitar, it would date the instrumentabout I790. The other guitars,fortunately,are labelled and dated.
SPANISH GUITAR BY RAFAEL ROLDAN.

danmehizo en Malaga,Aio de 1797, the last figure of the date in ink. Head, back and sides of rosewood. Flat rosewood fingerboardto 9th fret, remaindercovered by extension of belly wood. Rosewood figure inlay on this extension, and similar inlays between bridge and end of guitar on belly. Ebony bridge (no saddle). Ivory nut. Eleven brassfrets on fingerboard, five ebony frets on belly. Immature Spanish heel. Spanishhead with six pegs on each side, and bridge with six pairs of string holes. Indicates six double strings, and is stage intermediate between five double strings and six single ones, giving a date to this fashion.Pearldiamond-shapedinlays in rosette round soundhole. Two transversebars below belly, one above and one below the soundhole, as in the Cusworth guitar. No sign of lateralor fan barring,or residues
10

Bearsprinted label:Rafael Rol-

of glue from such bars.One must not, however, assumefrom this that fan barringwas then unknown, as one finds modem guitarsof Spanish luthierswhich for cheapnessweremade with two transverse bars only. However, this Roldan is obviously of good quality and one would assume that the maker incorporatedthe then most advanced barring method. It probably comes from the transitionalperiod when the new idea of lateral strutting had not been universallyadopted in Spain. Dimensions: Body length I7j' 26" Scale length Bout 9" Upper Waist 6ti Lower Bout II" 4" Depth I f Fingerboardwidth Total number of frets 17 Back is flat. Table is slightly arched,by affixing a bridge with an arched base. Contrasting stringing round edge of belly and back, and up centrejoin of back. Two-piece belly and back, as in the Cusworth Hall guitar. Not in playablecondition; tone cannot be estimated,but craftsmanship, wood qualitiesand thicknessesindicate that it will be of less volume of sound but far greaterrefinementthan that of the Cusworth instrument.
GUITAR BY FRANCISCO PAGES. The Pages family, established Cadiz in between I794 and I819, were the greatestguitar luthiersof their day. The eminent authorityEmilio Pujol, in his treatiseon nail and fingertip on technique, The Dilemmaof Timbre the Guitar,equates the guitarsof Torres and Enrique Garcia to the violins of Stradivarius, Pages, Guarnerius other makersof the Cremona school. This Pages guitar and bears the following label: Engraving of Violin and Bow crossed over a Guitar (bearing belly ornamentationalmost identical to the Roldan guitar described above) laid upon open music books. Wording: Fabrica Francisco de de Aio de I8I5 Pages,en la Rl. Ciudad San Fernando. (ast two figures in ink). FranciscoPages will be one of the family, San Ferando being a neighbouring town to Cadiz. Back, sides and fingerboard of walnut. Head faced with walnut. Walnut bridge, no saddle. Back and belly slightly arched.Flat fingerboard extends to soundhole, over belly. Seventeenbrassfrets on fingerboard. Eleven and a half frets cear of body. The fingerboardof this Pages appearsto be in two very thin layers,each of about A" thickness. The lower layer shows signs of the slots cut for frets, and the upper layer, of course, bearsfrets at the moment. The interestingpoint about II

the lower layeris thatit is evidentthatthisdid not extendbeyondthe to eighthfret;one hesitates try to removethe upperlayerof fingerbut seemsclear at board, the evidence that,asin the Roldanguitar, one time the wood of the belly extendedalong the neck for threefrets towards head.Thisextension thebellywood hasborneanedging the of of ebony which is still visiblein the lower layerbetweenthe eighth fretandjunctionof body andneck. The ribsandbellyof the Roldanarejoinedby a continuous of strip blocksof kerfing,but the Pagesbelly isjoined to the ribsby separate section.This is the earliest triangular examplethe writerhas seen of thisseparate as blocking,now generally acknowledged givinga freer, to Both guitars yet equallysecure support the edgeof the soundboard. havethe transverse barring supported long block-strips by gluedto the in ribs,whichat one andthe sametimeholdthe bars placeandprevent the ribs from collapsingunderthe pressure the barringat these of points.The backribbingis similarly supported. Two-piecebackand roundedge of belly.Soundbelly.Plainwalnutor rosewoodpurfling hole rosetteof sectionsof ivory tubing, inlaid into a composition circlewith contrasting wood stringing roundthe outsideand inside. Furthertriple stringingat edge of soundhole.Two transverse bars aboveandbelow soundhole, in Cusworthand Roldanguitars, as but alsothree lateral in theshape anelementaryfan: earliest bars the of example the writerhasseenof fanbarring a guitarbuiltin Spain. in Headbears twelve pegs, bridge six pairs of holes. Ivory nut. Slightly more matureSpanish heel. Dimensions:Body length I8'
Scale length Upper Bout Waist Lower Bout 25x 83s 6#^
Iof

Depth

4'

2' Fingerboardwidth The Pages guitar, which has been well used, is in excellent preservation and needs only minor repairsto cracksand purfling. It has been strung,and proves to have a sweet and mellow tone of greatrefinement.
GUITAR BY JUAN PERFUMO

Perfumo is listed by Prat (Diccionario de guitarristas)as a notable makerof guitarswho modelled his style andcraftsmanship the school on of Pages. This guitar bears the following label: Engraving of guitar (with a ribbon for slinging over shoulder) laid over a book of music. Label edged with engraved figuring. Wording of label Me hizo
12

al en Cadiz JUAN PERFUMO. Calle de Flamencos, salir a la calle No. I97. 1847 (all four figures printed). Nueva Head of rosewood, inlaid satinwood strip. Signs that head has borne an unknown number of holes for pegs. Now converted to machine tuning, with English machines, since they bear the Crown and the initials V.R. Scalloped bone nut for six strings. Original bridge is replacedby Germanpeg bridge. Rosewood back and sides; four strips of triple stringing laterally along the back, while the sides bear an equatorial band of triple stringing. Mature Spanish heel. Highly ornate large soundhole rosette of pearl arrows and crossesinterposed between circles of triple stringing in rosewood and satinwood. Rosewood 'Star' inlay on belly, between bridge and end. Two transverse and five lateralfan bars.This, the Pages, and the Roldan guitarsall have slipperdesign end-blocks at the neck end of the body. The guitaris in superb condition apart from the inappropriatebridge, and the tone, even with this bridge, is powerful and refined. Dimensions: Body length Scale length Upper Bout Waist
Lower Bout

i8?"
25'

9It 7|'
I2"

4*' Depth Ijf Fingerboardwidth Body joins neck at I2th fret I8 German Silver frets
PETERSEN GUITAR

The dimensions of this guitar, a modern concert guitar illustrated in Plate I, are: I9" Body length
Scale length
25S'

Io"' Upper Bout Waist 9lf Lower Bout 14r: 3X' Depth 2A" Fingerboardwidth 19 nickel silver frets of T-section Arm joins body at I2th fret One transversebar, and a complex system of lateral barring described, together with other internal details, in a later paragraph.
13

Back and belly appreciably arched. Back, sides and fingerboard of rosewood. Head faced with rosewood. Six machine tuning pegs. Spanishtype bridge. Mature, outswept Spanishheel as in fig. 4c.
CLIFFORD ESSEX GUITAR

This guitar by Marco Roccia (Plate II) has two transversebars and an empiricalbarringsystem, at various angles to the length, below the bridge. Further details are given in A. P. Sharpe, The Story of the SpanishGuitar(see bibliography). From the photographs and descriptions of these six instruments, together with the many others describedand illustratedby Sharpe,the of gradual evolution of the general characteristics the modem guitarra de concierto be traced.An examinationof the changesin individual, may detailedpartsof the guitarduringthe pastcentury and a halfis, perhaps, even more illuminating. WOODS AND THICKNESSES Generallyspeaking,similarwoods are used today for the best instruments as were used in the eighteenth century-hardwoods such as walnut, rosewood, mahogany, maple, sycamore and satinwood for the back and ribs, and pine or spruce for the bellies. If there has been any development in this direction, it has been in the universalacceptance of palisander(South American rosewood) as the finest wood for back and ribs. The practiceof veneering the ribs and back with a hardwood upon a softwood basehas been entirely discontinued.Rosewood and ebony are still unequalled for the fingerboards,and a variety of woods are used for the neck. One cannot recall having seen an eighteenth-century guitar with a laminatedneck though one believes they do exist, but there is an increasingtendency amongst the best luthiers nowadays to laminate the neck with woods of differingcharacter,not just for ornament but to guard against warping and to bear the increased strain of the heavier stringing and greater scale length of today's instruments. It is difficult to generalizeabout the thicknessesof wood used in the guitars of the past and in those of today. One cannot consider this questionwithout relatingthe wood thicknessesto the overall size of the instruments,and to the strainimposed by the differing string tensions, then and now. But one can say with conviction, that the belly-wood thicknesstoday is much less, in proportion to the overall size and the string tension, than it was in the eighteenth century, whereas the relative thickness of the back and ribs has remained fairly constant in proportion to the greater size and stresses.What has happened is that a new system of internal supportshas been devised which enables the
I4

comparativelythinner belly-wood to bear without collapse the strains of a wider belly and greaterstring tension. This subject will be developed under the head of Barring. SCALE LENGTH Guitars of all dates show a remarkable variation of scale length (i.e. the freely vibratinglength between the bridge saddleand the bone or ivory 'nut' over which the strings pass to the tuning heads). One imaginesthat individualpreferencehas forced luthiersto supply instruments of varying lengths to suit hands of varying sizes; but just as the keyboard instrumentshave finally evolved a standardoctave span and key width, so the luthiers of today have apparentlyagreed a standard string length. Guitarsof the past have varied in scale length from as little as 23" to as much as 25j": the Cusworth Hall guitar has a scale length of 24[". The luthier,of course,has to set against the addedconvenience and ease of handling of a short scale length, the loss of tone from shorter strings and the uncomfortablecrowding together of the frets in the higher positions. In the end, he appearsto have decided, wisely, that since the stringsof today will stand the strainof tuning to the presentconcertpitch at the longer scalelengths, the player with the small hands must either seek an older guitar or learn to stretch the webs between her fingers until she (for the small hands are usually those of a woman) can span the stretches imposed by a scalelength of 65 cm., or 26", the standardrecognized today as the optimum. Guitars of the late Victorian era found in Britain, incidentally, usuallyhave a remarkably short scale length: 25" and even less is quite common. Was this not probably so because the guitar at this period was of the drawing-room accomplishmentsof the elegant young womane FINGERBOARD AND NECK (fig. 2) In fingerboardsand necks, also, guitars have always shown a great variation.The luthiers of the Elizabethanage understoodthe necessity for a fingerboardof a width sufficientto enable a finger placed on, say, the second string, to remainthere whilst the first and third stringswere vibrated, and that the strings had consequently to be so far apart that they did not raspagainstthe adjacentfingers, though some of the width was needed to accommodate the double stringing. The eighteenthcentury guitar, therefore, usuallyhad a fingerboardof adequatewidth. But the Victorian age brought a deterioration;the instrumentwas less seriously viewed, was a social rather than a musical accomplishment, and the playersfell into slovenly techniques,such as the holding of the neck of the guitarin the crook of the left thumb. This position makes it impossiblefor the left hand fingersto stretchacrossa wide fingerboard,
5I

and the consequencewas that the young women of fashion demanded a guitar with a narrow fingerboard,which they could encompasswith their small hands and inadequate techniques. But Madame Sidney

c)

cL)

Fig. 2. a. Flatfigerboard. b. Scallopedfingerboard.c. Standardneck (section). d. 'Martin'neck with oval fingerboard.e. 'Mozzani' neck. f Head of guitar made by Georg Stauffer(Vienna, 1830) Pratten, professor of guitar at the Guildhall School of Music, London, who put her mark of approval to many guitars of the later Victorian era, notably those of Streiff Misner, must have exercised a goosl influence against these malpractices, for one has never seen a guitar bearing her imprimatur whose fingerboard is excessively narrow.

I6

Even so reputable a luthier as Charles Frederick Martin of New York, whose guitars are acknowledged the finest made outside Spain in the last century, pandered to this fashion by designing some of his guitars with a fingerboard width of just over il' and a neck whose section was a letter V, to fit into the crook of the left thumb. These instrumentsare very difficult to play by the acknowledgedly correct technique of today, in which the ball of the thumb and the tips of the fingers are the only parts of the player's left hand permitted to touch the guitar; the ball of the thumb finds the ridge of the V-shaped neck most uncomfortableand inconvenient. Other makers,such as Mozzani, went to the other extreme by flattening the back of the neck for the ball of the thumb to rest on, but this reduced its rigidity and is now seldom encountered. Happily, during the last decade, all reputableluthiers have accepted as standardfingerboardwidths a minimum of 2" at the nut end, and a maximum of 21", and this has disposed of the problem of the narrow in fingerboardwhich confrontedguitarists Britainin pre-second-worldwar days. The ovaled figerboard, often used by Martin and Panormo, has also largely been abandonedin favour of a flat-topped shape. The oval shapealso panderedto the less securetechniquesof the past, when the first finger could not be held straightin the grand barre. FRETTING (fig. 3)

In the early part of the nineteenth century there was a certainvogue, which persistedinto Victoriandays, to dispense with raised fretsand to cut and raspout the fingerboardin a seriesof waves, the crestsof which coincided with the positions in which the frets were normally found. In many instancesthis wave-form was combined with ivory or metal frets, which were inset into the crests of the 'waves' so as not to protrude beyond the surface,but to prevent, by their harderconsistency, the strings from wearing grooves in the ebony or rosewood fingerboard. These scalloped fingerboards,as they are named for want of a better word, are very pleasant to play upon, and the glissando and portamento, effectsthen much used in certainkinds of music, are easier performed than upon the normal flat, fretted fingerboard.But there is no point between the wave-crests at which the left-hand finger may trap the string, so that in difficult works the player has a sense of insecurity. The pitch of the note is definitely changed if the finger should happen to fall in the trough of the wave, so to speak;and there is excessive wear upon the wound basses because the strings are constantly dragged to and fro over the wave-crestsby the pressureof the
17

fingers. Scallopedfingerboards,therefore,have apparentlybeen finally abandonedby luthiers. The practice of veneering the back of the neck and the heel with ebony or other hardwoods is also now obsolete, probably because of the cost of such processes in these days of higher living standards amongst skilled craftsmen. It was a common practiceof the eighteenth century to fret the guitars with solid ivory or ebony frets of oblong section, with the sharpedges rubbed down where they encounter the left hand fingers. The frets were let into the fingerboardup to the body, but in most instancesthe

/li

9)

c)

b. nut. Fig. 3. a. 18thcenturyfret (section). Modernfret. c. Standard d. Imnut. proveddoublenut. e. Roccia-Essex fingerboard did not extend over the body, but ceased at this point. From the body junction up to the i8th fret, if fretted as far as this, the frets were also of ivory or ebony of squaresection, glued to the table of the instrument. The next development was the extension of the fingerboardin one
i8

piece past the junction of neck and body, right to the soundhole. This was an important development, since a fingerboardof ebony, often a quarter-inchthick, is a significant factor in strengthening the whole arm and upperbout of the body againstthe stringtensionsand in allowing heavierstringsand a longer speakinglength to be used. In the early years of the nineteenth century, the use of ivory frets gradually died were frettedwith oblong-section brassfretting, out, and the instruments let into the fingerboardto half the depth of the fret along the whole length from nut to soundhole. Finally, the tongued fret was introduced. This is drawn through a series of dies to the required shape in the form of a continuous wire; the upper flange is of half-oval section, with a thin tongue protruding from the lower surface for embedding in the fingerboard.This wire fretting does not have any sharpedges to cut the strings or abradethe fingers, and its exact method of manufacture enablesany fingerboard which is accurately surfaced, to be accurately fretted without the necessity for drasticrubbing-down afterwardsto level off the tops of the frets, thus preserving the half-oval smoothness of the fretting to the player's touch. There is still some controversy as to the best width for a fret: narrowerfrets are said to give more precise intonation and a clearer tone, but wider frets are claimed to give more volume and not to cut into the stringsso rapidly. One would prophesy that the wider frets will survive. The wide bone fret nearestthe tuning machines,known as the 'nut', has the dual primary purpose of holding the strings the requireddistance apartand the requiredheight above the first fret. There has from time to time been advanceda theory that no one nut can satisfactorily combine both functions and that two end frets should be provided to a guitar, one of ivory to space out the strings, and another of metal set immediately in front of the ivory fret, to hold the strings the correct distance above the fingerboard. It is true that there is great difficulty in so cutting the slots in the ivory nut that it will serve both purposes, but despitethis, the alternativemethod of bone and metal fret combined has never really secured wide adoption by luthiers. One can see no basic objection to the improved method, and suspects that natural conservatismis at the root of the objections (not a bad thing, either, in a makerof musicalinstruments).One luthier,Marco Roccia of Clifford Essex Ltd., hasingeniously combined the two types of nut by cutting from one piece of ivory a double nut, the back of which is the spacerbar, and the front the level-bar (fig. 3). This nut is most effective in practice, and is an advance on all other types. One wonders why it is not universallyadopted.
19

is methodof laying a guitarfingerboard so to make The standard andfit it thatits top surface an exactlyflatplanefromend to end,so is rise in inclined the strings that gradually fromthenutto thebridge such a slope that, no matterwhere fingered,the stringscan vibratewith maximumintensitywithoutrattlingupon the frets.Experiments are to now proceeding with slow-motionphotography confirma theory that the stringsdo not vibratewith consistent intensityat all points claimsthat M6nch,a leadingGerman luthier, alongtheirlength;Edgar he has provedthis theory,and is now buildingguitarswith fingerwhosesurfaces a complexseries minuterisingandfalling are of boards curvesso as to give at everypoint upon the figerboard the smallest with freedomfrom the strings possibleclearance compatible jarring the uponthe frets.But for all practical purposes, straight fingerboard one remains, believes,the mostsatisfactory. of The methodof settingthe positions the frets,at one time based rule of thumb, or upon the divisionsof the diatonicscale as upon accurate appliedto any one string-and reasonably only in a few now been superseded systemsbasedupon the adjacent keys-has by formulae applicationof harmoniclengths or upon mathematical which enablethe fretsto be set at regularly and regularly graduated distances diminishing apart,in orderto producea scalein equaltemwith the intervalsobtainedon an equallyperamentcomparable instrument. mechanical matheand tempered keyboard (Themethods, usedto secure resultaretoo lengthyto describe this matical, here,but well be deferred a laterarticle, to shouldthe subject of be they might
interest to readers of this JOURNAL.)

ATTACHMENT OF NECK TO BODY AND HEAD (fig.4) Themethodofjoiningneckto bodyis of somesignificance only not in the durability the instrument also in the qualityof its tone, of but sincethe stringtensionpullsthe headtowardsthe body andthe body towards head,andas the hardwoods neckandribsarenot comthe of the triesto fold up, with the hinge of this pressible, whole structure movementat the spotwherethe neckandbodyjoin. Evenin the best there is an appreciable instruments movementabout this axis when the stringsare broughtup to tension,so that they then lie a few thousandths an inch higherfrom the fingerboard of than when the is instrument first strungslackly.Where there is any defect in the methodof securing neckto body,or wherethe workmanship faulty, is the weakness showsitselfover the yearsin a gradual movementof the neckin its angleto the body,so thatthe stringheightabovethe finger20

boardbecomesprogressively and greater, the bridgesaddlehas to be loweredto compensate-withdetriment the toneandvolumeof the to instrument since the leverageis reducedwhen the bridgeheight is
reduced.
2

//^/

j"""/////f/

ENDaJ
|-'

BLOCK

b/

SECTr OhN

I.

\-|P|? i1111

ENDBLOCK

i!!!il

(of noArrow oLA

-]BO

Fi.4.ozn' hl bFSLIPPER
heel. b. French heel.c. Spanish heel. Fig.4. a. 'Mozzani' Guitarsmade in Spain, or to the Spanishpattern, differ at this point from those made to the Scandinavian,German or French and Italian patterns.In the latter, in many instances,the body is built with endblocks at both ends, similar to violin end-blocks. The neck is then attachedto the body by a flat gluedjoint, or less frequentlyby a dovetailedjoint, especiallyif the neck is detachable,in which case the joint is not glued, but is secured by a threadscrew and key which passes, throughheel andend-block.If the neck is detachable,the fingerboardis
21

securedto the neck but free from contact with the belly of the guitar, so that it may not rattle when the belly vibrates. If the neck is permanently secured,the fingerboardis glued to the belly as well as to the neck, giving a more secure structurethan the best detachabledesign can afford.On the other hand, makersof the Spanishmodel frequently make the heel and end-block in one piece (or in a series of horizontal layers like plywood) and the ribs are glued into slots in the neck-heelblock ensemble. This method gives greater rigidity than the method first described,but it is not so easy to set the neck at exactly the correct angle, and if it becomes necessary,by reason of warping or breakage, to fit a new neck, the operation is much more difficult. Consequently makers are now in some instancesbuilding to the Spanishmodel, but affixing the neck by the French method. The majority of makers are now making their end-block at the neck end of the guitar in the 'slipper'pattern,with a 'tongue' extending along and glued to the back of the guitarfor three or four inches, and sometimes a second 'tongue' extending along the underside of the belly below the fingerboard. These tongues give extra rigidity and resistanceto the tendency of the instrumentto collapse at this pivotal point. The heel, extending downwards from the neck towards the back of the ,exteriorof the ribs is found in a number of designs, of which the Spanishdesign is the most aestheticallypleasing as well as probablythe most efficient in meeting its share of the stresses.A few makers, such as the Guadagninisand Mazzani, dispensedwith the heel altogether. The Italianluthiers commonly make neck and head in one piece. To avoid the use of excessively large blocks of wood from which to carve the unit, and also to avoid the grain of the wood in the head running too sharplyacrossthe line of the head, they make the angle of the head to the neck less acute than when the head, as is the common practice, is separatelycut and afterwardsspliced to the neck. BRIDGES (fig. 5) The bridge of a SpanishGuitar is always glued to the belly of the instrument; there is no separatetailpiece. But a significant change in bridge design occurredtowards the end of the eighteenth century. The bridges upon eighteenth-century guitars were basically (disregarding the ornamentalside arms, which often extended right acrossthe belly) a rectangularblock of hardwood of the general shape of a railway sleeper, through which, from back to front, were drilled holes for the strings at approximately half the total height of the bridge above the belly, one hole to each string. This method gave the player a rather arbitrary choice of two string-heights according to the manner of
22

securing the strings to the bridge. The first method was to pass the string through the bridge from soundhole side to tail-end side and there to knot the string: this gave the string a bearing on the lower side of the drilled hole, at half the height of the bridge, and the lay of the fingerboardwas so arrangedthat this position kept the stringsvery close to the frets for a player of weak touch, e.g. a woman or a casual

O<^EB^BC)^Oe,

, ////

///!

C.,h~
C)

mLm~

f) /,//

??~
////

...;

??
I/ / /i

9)

)
4)

9
Fig. 5. Bridges: plans a. I8th century b. 'Panormo' c. 'Lacote' d. French

///

Sections e. I8th century (entirelyebony or rosewood) fandg.'Panormo' (ebony or rosewood) h. Lacote' (ebony body, ivory saddlefillet) i. Modem 'Spanish'
23

player. For the professionalor habitualplayer with a strong left hand, the second method of securing the string was used: the string was passed through the bridge from tailpiece side to soundhole side and there knotted: the free end was then passedover the top of the bridge and thus over the fingerboard up to the tuning peg. This gave the string a bearing upon the upper surface of the bridge, and thus the stringslay furtherfrom the fretsso that they could be struckvigorously without causing them to rattle on the frets as they vibrated. But the system had the disadvantagethat as the belly and the bridge twisted with the constant string pull, the string ceased to bear upon the front edge of the drilledhole, or upon the front edge of the bridge-top as the case might be; the bearing thus became indeterminatelyapplied to the whole width of the bridge, in which event the string buzzed and rattled, or it even came to be applied at the back edge of the bridge, which completely upset the intonation of the instrument. Around the turn of the century, bridge design was accordingly modified by the bridge being made much wider from back to front (with consequent added security of attachmentto the belly) and the oblong bar which had been the whole bridge was then transferred the tail-end or back to of the bridge, so that there was room at the front or soundholepart edge of the bridge for a fillet of bone, ebony or ivory to be inserted in a groove, the fillet forming a firm bearing for the strings which pass over it. This style of bridge is known as the Spanishbridge. The bridge design was of course then so arrangedthat the drilled holes in the back bar were much lower than the upper surfaceof the bone fillet, so as to give considerabledown pressureof the stringsupon the fillet when the stringswere undertension. The principaladvantageof this bearingbar, apart from the securebearing, was that as the guitar settled down and assumedits final neck angle and belly position (all guitars'move' a little during the first few years of their life) the height of the strings could be changed to suit the new positions, or to accommodate individual tastes in string height, simply by substitutinganother bone fillet of a differentheight. A variantof this bridge is the so-called Frenchbridge, in which the bone fillet is as in the Spanishbridge, but there is no back bar; instead, holes are drilled through the bridge and belly of the guitar,one to each string, and the knotted ends of the stringsare secured into these holes by the insertion of bone, ivory or ebony pins or skewers. In instrumentsbeing built in the mid-twentieth century, the Spanish bridge has supersededthe Frenchbridge. One might add that today's ornamentalshapes, and apartfrom bridgeshave abandonedsuperfluous occasionalcolouredwood inlaysaresoberlyrectangular blocks designed
24

PLATE I

Left to right: The Cusworth Hall guitar, c 1790; Pages, I815; by Perfumo, 1847; and b

PLATE II

Clifford Essex Guitar, I953

lReproduced froml A. P. Sharpe, The Story of the Spanish Giuitar, by permission of the Clifford Essex Music (Co.

to perform their function in the most efficient way possible. The size of bridgeshas increasedin conformity with the generalincreasein body sizes, but the weight, proportionateto the areaof the belly, remainsthe same: it is the thinly cut 'arms' of the bridge which have been extended, not the solid body of the bridge. ORNAMENT It is difficult to deduce any logical pattern in the changes in ornamentation of guitars during the period under review, since there has always been such a wealth and variety of ornament in applique hardwoods upon the belly as part of the bridge design, in rosettesover and around the soundhole, and in inlays on ribs, back and head. One would say, however, that with the increasingsobriety of male attire,the guitar has graduallylost its adornments;would there not be a connection as well as a parallel? The ornamenting of guitars, particularly with mother-of-pearl, reachedits zenith in late Victorian times. The writer de owned at one time a guitarra concierto Antonio Carlos Garcia,one by of the few true concert guitars made by that maker, who normally made instrumentsof a cheaper grade, with flat bellies, for casual use. This A. C. Garciaconcert guitar was of the largestconcert size and the ribs and belly were almost completely covered with ornate mother-ofpearl inlay of superb craftsmanship-but the total effect was in the most dubious taste. As the guitar in recent years has become more and more seriouslyregardedin musicalcricles,so has its ornamentation dwindled until one now finds that guitars by the best makersbear no more ornament than an inch-wide band of purfling of inlaid coloured woods round the soundhole, and stringing in a contrasting colour at the edges of the table, ribs and back. STRINGS The greatestdisadvantageof the Spanishguitar until the presentdecade-a disadvantageso severe as to deter most people from taking up the guitar as a professionalcareer-was that the gut treble strings,imperative (as distinct from steel strings) for producing a mellow tone suitable for the interpretationof serious music, would not stand in tune, were frequently false in vibration and so could not be accurately tuned. They constantly broke, and even when a perfect string was fitted and manoeuvred with much patience into proper tune, it would not be many days, or even hours, before the gut began to fray and abrade under the nails of the player and the hard fretting of the fingerboard. Moreover, the lower stringsof deeper pitch, which must be wound 25

with copper wire to increasetheir mass, were wound upon silk floss, the only suitablematerialthen available,and this floss, if it were to bear the tension of modern concert pitch at an adequatescalelength, had to be of such thicknessthat the resilienceand sostenuto of the string were detrimentallyaffected. The development of synthetic materials,notably nylon, has changed the whole situation.Monofilament nylon treble stringsare now available, accuratelyextruded and of constantdiameterthroughout, which give an absolutely true note, will stand in tune indefinitely, and will bear the hardestuse over a period of many months without wear or breakagethough the tone gradually deteriorates.The bass strings are now wound upon a core of nylon floss, which has many times the breakingstrainof silk floss, and the cores of the bass stringsmay now be made so thin that the strings can vibrate freely, with ample (and sometimes too much) sostenuto. Other materialsare now being tried for strings, and the writer has used a Luron first string, with encouraging resultsof clearertone and greatervolume. Doubtless as these synthetic materialsare perfected in greaternumbers,the ideal one will emerge. Meantime, such is the suitability of nylon stringing that the guitar is now accepted as a fully practicableinstrumentfor a professionalcareerin music, and is offered in the Royal Colleges of Music in Britain as a second study when desired. VARNISH There has been no significant change in either the materialsor the methods of varnishing fretted instrumentsduring the period under review. The question of varnish and its application to the guitar and mandolin was exhaustively reviewed by the American experimental luthier Harry Garmont in the B.M.G. Magazine (see bibliography) in the issuesof Februaryand March I952, and by the British authority John W. Duarte in the GuitarReview (New York) No. 12 (I95I). Brushed on oil varnishover a sealing coat of white of egg is favoured by many leading makers. Others use french polish, rubbed on. MEANS OF TUNING As the watch-key-and-lever type of tuning has apparentlyalways been confined to the cittern and the Portugueseguitar, one may conclude that there have been only two traditionalmethods of tuning the Spanishguitar: by wooden pegs (as with the cello) or by tuning machines (aswith the double bass).These two methods arestillin common use, and there has been no significant change except in emphasis: at the end of the eighteenthcenturynearlyall guitarswere tuned by pegs,
26

de are whereas today, nearly all guitarras concierto tuned by machines, are and only the guitarrasflamencos tuned by pegs. (The flamencoguitar is a lighter-built instrument, usually with back and sides of apple or pear wood, made to produce a brilliant, hard tone for accompanying flamencosingers and dancers.) Pegs are used on the flamenco guitars, one concludes, not only becausethey are cheaper (the flamenco guitar is an instrument of the people) but also because they are lighter in weight; the heaviermetal machinetuning head would not balancewith the lighter weight of the body of the flamenco guitars, and the head would tend to fall when the instrument is balanced upon the performer'sknee. The style of head favoured on the Continent, shown in fig. 2f., with all the tuning buttons on one side and with the machines enclosed by a back plate (usually engraved), has died out; its disadvantage was that the buttons were too close together. BELLY DESIGN The writer has not yet seen a guitar built before I8oo with a belly which has been built upon the principle of the arch, so as to add to its strengthwithout increasingits weight or detractingfrom its vibrancy. But today, all true concert guitarsare built with an arched belly. The arch is impartedpartly by cutting the belly wood in two pieces with a join down the centre so shapedthat when the edges are pulled together the wood tends to form itself into an arch:partly by gluing lengths of spruce or pine known as bars, or struts, to the undersideof the belly whilst this arch is still maintained, so that when the glue is dry the archis permanentlysustained;and partly by thoroughly drying out the wood of the belly during the whole of the manufactureof the guitar, thus shrinkingit as much as one per cent-when the instrumentis then taken into a normal atmosphere, the shrunk wood expands to its previous size, and the pressurehelps to increasethe degree of arching or to maintainthat arch which has alreadybeen impartedto the belly, despite the twisting pull of the stringsupon the bridge which tends to collapse the arch. The sole of the bridge is of coursearchedto fit the belly, and this arch of the bridge is another factor in maintaining the arch of the belly. Some luthiers, particularlythe Italians,save themselves the trouble of these complicated processes by cutting the ribs to such a shape that where they join at the tail end, they graduallywiden so that when the belly is attached,a slight ridge or false arch is formed in the belly. But though this performseffectively one of the functions of the true arch, i.e. to stiffenthe belly without adding to its weight, a false arch (which is to a true arch as the canopy of St PancrasStation is to the dome of
27

St Paul's) hinders the vibration because the stiffness is all along its length, instead of being at the region of the bridge only, where the stiffnessis required. The soundhole remains circular and in approximately the same relative position as it has always been; it is only in the plectrum guitar, or jazz band guitar, that the cello type of belly construction,with bass bar, has been evolved to produce, with steel strings, the hard, cutting tone required for the harpsichord-like 'continuo' rhythm in dance orchestras. The plectrum guitar'ssteel stringsare, of course, struckwith a tortoiseshellplectrum, whereas the strings of the Spanish guitar are never of steel, and are struck with the fingers of the right hand. INTERNAL SUPPORTING BARS (fig. 6) is in this field that the Spanishguitarhas seen the greatestadvances, It if one may thus describe them, since the eighteenth century. These advances are not merely an improvementin the internal design; they representa revolution in the conception of the purpose of the bracing bars within the body of the instrument. Guitarsof the first half of the eighteenth century were invariably bracedby barsof spruceor pine-the same wood as the belly-running transverselyacross,and glued to, the bellyparallelwith the line of the bridge and at right angles to the line of the strings.It was customaryto place one or two such barsbetween the soundhole and the joint of the body and neck, and either two or three bars between the soundhole and the lower end of the body beyond the bridge. Lesscommonly one finds an instrumentwith only one bar to this part of the belly, and that placedabout one inch beyond the soundhole,between it and the bridge. The back, too, was braced by two, three, or four transversebars, sometimes of pine but sometimes of the same wood as the back and alterationin the design of the sides,but as there has been no appreciable barringof the back, this aspect may be ignored for the purpose of the presentinvestigation. The below-belly barring of the eighteenth-century guitar had but one purpose:to stiffenthe belly againstwarping and againstthe torsion of the string pull as applied to the bridge, which constantly tended to depress the belly between bridge and soundhole and to elevate the belly between bridge and tail end of the instrument.If the belly wood was made thick enough to withstand this pull, the tone of the instrument was hard and brilliantand without depth of bass notes; in other words, the upper partialsin the tone were favoured by the rigidity of the belly. Only by making the wood of the belly very thin, could depth of tone be impartedby adding the lower partialsto the tone. But when
28

this was done, the belly soon twisted, and the brilliance of the treble tone was lost becausethe belly was 'soggy' insteadof taut. To counteract this, and to provide a belly thin and resilientfor good basstone but also taut and vibrant for good treble tone, the luthiersnot only placed a transverse just beyond the soundhole (which held the waist firmly bar

')

K00

L-----------

0
\4) ^

(*NO T PAkSVRt.
~RA94SVtSE oAR.
CLUEO tr SIELLY

(\oI 0

(
\/ \ \ /\D\DD/ \\

0I 11 O\JLY. \\ \\ \

CCNtE

Fig. 6. Barring below belly a. I8th century. b. 'Panormo' in the Spanish style. c. C. F. Martin. d. Antonio Torres. e. Extended system.

and prevented complete collapse) but also another transverse just bar beyond the bridge and, if the body was large or the belly wood very thin, a third between the soundhole bar and the furthest strut. The appended diagrams show the common positions of these bars. It will be noted that the purposes of these bars were merely to support and tauten the belly and to hold the ribs in position. The barsran transverselyto the length of the body and, unfortunately, tended to
29

to since the carry vibrations the sideof the instrument, thosevibrations follow the line of the ribacross belly.The barring, the therenaturally and to madeno fore,otherthanaddingtautness strength the structure, to contribution the volumeof tone of the instrument'. It is difficultto fix a date to the discoveryby luthiersthat these bars of as transbracing werenot merelycapable support whenaffixed the of verselybut, if affixedlaterally,could transmit vibrations the to bridge(carried the belly at the point wherethe bridgewas set) to of had areas thebellywhichpreviously vibrated onlyfeeblyandwhich, in fact,had functioned as the completionof the enclosing walls only of the amplifying soundbox whichis the body of a guitar.It is known, a however,thatin the latterhalfof the eighteenth century certain type of guitarbecameknown as that 'madein the Spanish style', and as
Louis Panormo2about I8Io adopted this new method of barring and changed his guitar labels to 'Louis Panormo, Maker of Guitarsin the Spanish Style', we may assume that this fundamental discovery was made in Spain towards the end of the eighteenth century. The essence of the new method was to distributethe vibrations over the whole of the lower bout of the guitar belly (the larger of the two circles of the figure-8 wliich is the guitar shape) by a lateral fan of bracing bars which replaced one of the customary two transversestruts between soundhole and tailpiece. The focus of the fan was between bridge and soundhole, and the bars spreadout towards the bottom of the instrument so that the vibrationswere evenly distributedover the lower belly

There were, of course, many variantsof this method: the number of fan bars varied from three to nine, and is still not standardized.Some luthiers, Charles Frederick Martin notably, substituted a cruciform interlocking bracing bar for the fan of struts (fig. 6c). This produced a brilliant and powerful tone, but imparted a certain nasal quality such viola tone from violin tone, and which is not considered as distinguishes to be the authentic Spanishguitar tone: this form of barringis still to be found occasionally in the instrumentsof today, especiallyif cheapness of production is desired, but its use is dying out. Just as the design of the violin reachedits zenith in the instrumentsof the Cremona school, so the design of the guitarappearsto have reached at least the first stage of perfection in the instruments designed and made by the Spanishluthier, Antonio Torres, in the second half of the nineteenth century (fig. 6d). Torres' instrumentsepitomized the best in guitar design of their day. To the customary seven fan bars of the Spanishstyle he added two further bars semi-laterallyat the tailpiece end of the belly, and other bars and plates in the region of the sound30

(fig. 6b).

hole, which gave the instrument an unmistakabletonal characterof mellow sweetnessand power which no other design can reproduce. It is not the purpose of this articleto examine the many examples of unusualguitar design which have been evolved, but ratherto examine the mainstreamof guitar design which has culminatedin the standard de of guitarra concierto today as used by AndresSegovia and other leading professionals.Iron bracing bars have been tried; double soundboards; detachable necks; false backs; 'sound chambers' and 'sound funnels' within the body of the instrument;additionalbassstrings;lute guitars, harp guitars, lyre guitars and what you will; but the professionalof today usesconventionally-shaped six-stringconcert guitarsbased either upon the standardSpanishfan barringsystem or upon the modified fan barring of Torres. The standardbarring tends to produce a brilliant and lively tone, and the Torrespattern a more subtle and sweet tone of equal carryingpower but slightly less apparentbrilliance.Playersmake their choice from personalpreferenceratherthan becauseeither pattern is noticeably superior. However, the barring system of the guitar is still the subjectof much research,and there has been some advance in the presentcentury in the constantsearchfor greatervolume and body of tone for concert work, especiallywith orchestras,where the tone of the guitaris too easily swallowed up by the generalmass of sound, and composers of concerti are correspondingly limited in their scope to antiphonal treatment of the solo instrument or to works with very small orchestralgroups. These advances have come partly from researcheswith modem scientific instrumentsand by empiricalmethods into the vibration properties of woods, so that the luthier is able to choose the most suitablepieces of wood for his instruments,and partly from a new conception of the function of the belly of the instrument and its barring.Luthiersand designersin the forefront of researchnow recognize that the best resultswill come from a conception of the body of the guitar as a firmly built box whose back and ribs are of themselves sufficientlystrong to bear the majority of the stressesto be imposed upon the instrument. To complete the sound chamber, instead of mounting upon the back and ribs a belly whose wood is thick enough to contribute to its total strength, the luthier now attemptsto stretch over the top of the body a thin skin of wood which has very little strengthin itself, but is supportedby a web of barswhich perform the dual function of taking the twisting stressof the strings as applied through the bridge and of transmittingthe vibrationsof the stringsto whatever part of the belly the bars may extend. The first step was of course the adoption of the 'Spanish Style' of construction, with a fan of bars supplemented by several transverse
31

bars.Thiswasimproved Torres the addition two further bars of by by at the tail end, andplatesandbarsby the soundhole. About I935 the writersawa guitarmadein BuenosAireswhichhadadvanced beyond thisdesign.It retained transverse by the soundhole, the bar the bars but betweensoundhole bridgewas arched thatsomeof the fanbars and so couldpassbetweenit andthebellywithouttouching transverse the bar. Thisextended area vibration farastheloweredgeof thefingerof the as was board,andthe tone of the instrument very satisfactory. The writerhas for the past ten yearsexaminedand experimented with guitars, a alwayswith the objectof discovering methodwhereby the lateral couldbe completely since abandoned, thisis a barrier barring to the conveyingof vibrations alongthe belly. One methodwould be to mounttheselateral intothe sidesof the guitar aninchbelow bars half of the underside the belly so that the belly is entirelyfree of contact with them,andso thatthe fanbarring be maytherefore extended past the soundhole the freeareas the bellyin the upper(smaller) to of half of the figure-8whichis the guitarshape.The difficulty hereis thatthe is in to pull of the strings exerted two ways:firstly,the strings attempt the bridgetowardsthe soundhole, secondly,they attemptto and pull the pull the heckof the guitartowards bridge,with the effectthatthe tendsto be elevated its headportion,and depressed in in fingerboard thepartbetweenthejoint with the body andthe soundhole. transThe versebarjust by the soundhole, beneath terminal and the point of the is placedthere to preventthe belly being sharplydefingerboard, here of withoutthis pressed by the pressure the endof the fingerboard; bar, the body would graduallycollapse.The problemis to support the lowerend of the fingerboard (whichis, of course,gluedto the top of surface the belly) without interfering with the free passage the of fanbarring the fingerboard the upperboutof the guitar into alongside with both the transverse normallyglued to bars belly,and to dispense the belly in this region without addingto the total weight of supThe whole of the belly of portingwood within the upperchamber. sucha guitarwould be set in vibration when a stringis struck. THE FUTURE OF SPANISH GUITAR DESIGN of With the carrying the fan barring right to the topmostpartof the belly, the conventional designof the Spanish guitarwould reach finality except in minor detailssuch as the proportionsof wood and thicknesses the exactdimensions the variousparts.Most of the of modernluthiershave been content to copy the designsof Torres, Ramirezand otherearlier makerswhose designsare EnriqueGarcia, as standard; have been concerned accepted they only to make small
32

bassbar of placedby a 'treblebar' of similardesignto the familiar cello design.Steel-strung for use in danceorchestras madeto guitars this designhave been remarkably successful. an instrument If were builtto a similar butwith modified wood thicknesses respond to design to the lighterstrings the Spanish of it guitar, mightbe the idealinstrumentfor concerted with playing.The writeris now collaborating the American luthier who hasalready constructed revoa HarryGarmont, lutionarymandolinto this patternof cello design,and who is now the building firstexperimental Spanish guitarto cellodesign.Garmont is well knownfor his researches the properties wood asapplied into of to the buildingof musicalinstruments; usesthe finestand latest he scientific for of apparatus the studyof the acoustical properties instru33

secure the requisitesostenuto, the soundpost of the cello would be re-

suchas wood thicknesses. Britishluthier,however detailchanges One -Marco Roccia, who makesguitarsfor CliffordEssex Ltd., with whom the writerhas been privilegedto collaborate-recommenced afterthe secondworldwarby castingasideall preconmakingguitars ceivedideasexceptthoseof body sizeandshape,and,callinguponhis in of somethousands old guitars all the world's experience repairing by includeddouble makers, beganagainon new lines. The experiments cruciformand other unorthodoxbarrings;barringbased on that foundbeneath soundboards earlysquare the of withpianos;barrings out uniformity structure, of wherethe soundboard placedempirically wasprovedto be understress; to hollowneckswith openings the body andto the headend of the neck;andvariations wood thickness of on trebleandbasssidesof the belly andsides.One cannotsaythatany of these experiments produceda guitar superior,overall,in tone or volume to the orthodoxinstruments, they did produceguitars but whichwereparticularly suitable particular for purposes-e.g.forhome for playing,for song accompaniment, the playingof lute music,for and and of broadcasting recording, the like-and the selection designs to available the purchaser him an opportunity exercise own to his gave tastesand inclinations, particular counteringthe tendencytowards of greater uniformity tone whichmarksthe concertguitarsof today. Further if any, would thereforeseem to lie in an entirely progress, to The revolutionary approach the problem. writerhaslong advocated research the buildingof a Spanish into guitaruponthe general pattern of the body of the violoncello,not with any confidence suchan that instrument would necessarily superior generaluse, but because be for it mightresultin a Spanish of suchpowerandtonaldepththat guitar the guitarmightat lastmeetthe otherorchestral instruments terms on of reasonable when usedin chamber worksandconcerti.To equality

mental sound chambers, and for the choice of woods and the determination of wood dimensions.If, therefore, the instrumentnow under construction is not a success (and difficulties of right hand technique may arise from the use of a cello type bridge with the strings high above the belly) it may be assumedwith some certainty that the now fully developed conventional Spanish guitar represents the furthest point of guitar design which can at the moment be envisaged. ELECTRICAL AMPLIFICATION As the structureof the instrumentis not substantiallyalteredwhen a guitar is electricallyamplified, one has not dealt with this method of increasing the volume of sound of a guitar for concerted playing. are Aesthetically,there seemslittle objection: harpsichords now habituso amplified.The criterion to be applied, one would say, is: Does ally the extentof the amplificationso change the nature of the sound that it is no longer delicate and tenuous to the hearer,but becomes akin to an electronic organ? If so, the amplificationis incorrect. If not, and it merely allows the guitar to be audible, without markedly changing the tone colour, it is surely acceptable. One hopes, however, that other means may be found to solve the problem of audibility, as acoustic tone is always subtly more pleasing than electronic tone. ACKNOWLEDGMENTS The writer wishes to acknowledge the help he has received in compiling this articlefrom Mr Eric Halfpenny;Mr A. P. Sharpe,in reading the draft and.supplying photographs; Prof Miguel Alboniz, Milan, in giving information about Italian instruments and general guidance; Mr Walter Dobson, Mr Edward Doherty and Mr FrancisMiller in lending guitarsfor examination; and Mr Albert Birch, in making the drawings and diagrams,in lending referencebooks otherwise difficult of access,and in closely collaboratingin the compilation. NOTES 1Duringthe eraof transverse makers musthaverealized limitathe barring, for morebrilliant the bass tionsof thesystem, theytriedto makethe treble and at more resonant settingthe transverse by barring anglesotherthanthatof a with some smallbenefitto the tone in to the line of the strings, rightangle someinstances. s The guitaristJose Ferdinand Macario I780-I839, in collaboration with Sor, and LouisPanormo Rene-Fran5ois Lacote(one of the Mirecourt the luthiers instrument was for makers) responsible manyimprovements groupof French and of construction design guitars hisera,butalthough of in thedetail Panormo fan-bar the remained faithful the transto adopted Spanish style,Lacote always
34

choiceof wood andsuperlative versebarring, relyingon careful craftsmanship For his accountof this triumto produce fine guitars. the most authoritative virateyet written,see Sharpe(seebibliography). BIBLIOGRAPHY TheGuitar Review,published the ClassicGuitarSocietyof New York by on lutes,strings, (409East5othStreet,New York 22). Manyarticles guitars, ofluthiers guitarists; and and varnish related subjects; biographies reproductions etc. and of paintings, of all periods, showingguitars lutes;etc. razonada laguitarra,vol., 1934-55; de EmilioPujol,Escuela further volumes 3 An treatise the construction, on in preparation. exhaustive and technique music of of the Spanish guitarand vihuela,by the Professor the Guitar, University of Lisbon. in Buenos Thedilemma Aires,I934. A shortcritical oftimbre theGuitar, survey on of the schoolsof nailandfingertip tone-production the guitar. The A. P. Sharpe, Story theSpanish London: Clifford Essex& Co., Guitar, of of with of 1954.A summary the historyof the instrument, biographies past and makers composers. andpresent players, B.M.G.published Clifford Essex& Co., London(editor,A. P. Sharpe). by devotedto frettedinstruments containing and A monthlymagazine frequent its on musicandtechnique. articles the guitar, players, T. M. Hofineister 'A briefaccountof the luthier Torres',in TheGuitar Jr,
Review, No. I6, I954. With detailed sketchesof the constructionof Torres's

instruments. and London:Schott& Co., I954. PhilipJ. Bone, The Guitar Mandolin, of and Biographies somecelebrated players composers. of makers composers. and teachers, directory guitarplayers, biographical Note(seep. 9). Sinceauthorities in assigning dateto the addition a a of vary it to sixthstringto the guitar, wouldbe moreaccurate referat thispointto the of construction guitars of abandonment luthiers the habitual by havingfive thanto the introduction the sixthstring. of doublestrings, Musical CanonGalpin,European Instruments, Ioo, saysthat 'the change p. to of was fromthe five pairs strings the six singlestrings cominginto voguein But the latterpartof the seventeenth The century'. A. P. Sharpe, History the of that'it wasJohann Gottlieb Guitar, 9, states Naumann, p. Spanish guitartutor Duchess Weimar(1739-1807) suggested addition the of who to Amelia, the of to and the maker E, sixth,bass string theguitar' quotes bookby theinstrument Otto in support thisview. In thisbook, Ueber Bauund of den JacobAugustus in Halle,Otto writes'ThelateDuchess Ameliaof Weimarhavingintroduced into the guitar Weimar I788 ... a greatdemand theinstrument in for rose.... I mustheretakethe opportunity observe the guitar to that had originally only five strings. The late HerrNaumann, at ordered the Kapellmeister Dresden, firstguitar with the sixthor low E string, whichI at oncemadeforhim. Since that time the instrument alwaysbeen madewith six strings,for which has
35 der dieErhaltung GeigeundallerBogeninstrumente, publishedin 1817 by Reinecke Buenos Aires, I934. An exhaustive Domingo Prat, Diccionario deguitarristas,

Whilstit maywell be HerrNaumann.' its have improvement admirers to thank the end of the eraof five of six singlestrings thatNaumann's advocacy spelt of doublestrings,one is inclinedto concludethat the attribution the sixth since the stringto Naumannalone may referto its adoptionin Germany, in Review,New York, No. 8, I949, authorityLuis Elorriaga, The Guitar its mentions earlier adoption.In his article,'The Guitar,a concisehistoryof an and its origins earlydevelopment', the appears illustration bearing following with 5 doubleandone singlestring: fromthe Syntagma strung caption'Guitar, i6th century'. also Musicum MichaelPraetorius, by Elorriaga says'the celeaddedthe fifth string and bratedmusician poet, VicenteEspinel(155I-1624) to the guitarbut to dateno one knowswho addedthe sixth nor when that occurred. systemof important improvement Espinel... fixed the ingenious writes (p. 9), 'for a whichis stillbeingused'.But against this, Sharpe tuning credited VicenteEspinelas the guitarist who addedthe long time historians has fifth stringto the four-string that guitar,but laterresearch revealed Berof mudo,writingin I544 (eleven yearsbeforethe birthof Espinel) speaks the A footnote to this statement readsas follows: 'Besides five-stringguitar'. was novel The playingthe guitar,VicenteEspinel the authorof the Spanish which Le Sage afterwards re-wroteas The Esquire, Obregon, Life of Marcus adventuresGil Blas'.The chiefauthority attributing invention the for the of of was was fifthstringto Espinel the dramatist Lopede Vega, andhis statement Doisi de Valesco(I640)andGaspar Sanz(1674)in laterendorsed Nicholas by on theirtreatises the guitar. In Russia,duringandsincethenineteenth a was century, seven-string guitar and sixthe standard of instrument, amongstthe manyvariants the standard with havebeeninstruments guitarsincethe eighteenth stringSpanish century bass which did not extendover the fingerboard which and additional strings a separate, extended, or weremounted at the bodyend,andupon upon bridge headorsecond neckattheheadend:these wereplayed a second additional strings in basses keysfor whichthenormalopenbasses E, A of 'open'as fundamental A andD were unsuitable. combination harpandguitarwas also evolved of at the beginningof the nineteenth are century.Many of theseinstruments and in described illustrated Galpin in various and articles TheGuitar in Review, New York.But all haveeventually beenabandoned favourof the standard, in instrument. tone andvolumeof the guitararealways The six-string Spanish or affected, its lengthof life is seriously reduced, the addidetrimentally by of and instrument tionaltension theextrabasses, thelimitations thesix string of limitsof the performer, or arranger lie largelyin the technical using composer rather thanin its havingonly six strings. One would the standard instrument, will remainthe standard that six strings in prophesy pattern the foreseeable future.
TERENCE USHER

rectalistand broadcaster, of composerand arranger works for guitar;also sectionof TheGuitar editorof the Academy Review New York.-Ed. of
36

is Tutor of the Guitar, Royal Manchester College of Music,