Talk at the University of Indiana - 1973
Restoring and Reconstructing
the Harpsichord

Frank Hubbard

[Edited by B.J. Fine]

I see by today's program that I am expected to address you on the subject of restoring and reconstructing the harpsichord. This title, of course, has two meanings: the first would be the reviving of the harpsichord, the making it available to contemporary musicians and the returning it to use. The essential nature of the instrument would have to be rediscovered and those observations implemented in wood and wire. The second interpretation of the subject would involve the repairing or rebuilding of existing examples of old harpsichords. I plan to deal in some detail with both the revival and the restoration of the harpsichord.

Since the revival of an obsolete but accessible technology is largely a matter of motivation - the obstacles are easily surmountable - I can perhaps best define my own motivation by giving you an account of how it came into being.

In 1946 I had been newly discharged from the army after three dreary but fortunately unadventurous years, and largely motivated by a desire to take up my life where I had last dropped it, was enrolled in the Graduate School at Harvard, where I proposed to continue my study of English literature, a field I now realize aroused in me only a tepid interest. It wasn't the literature itself. I enjoyed that well enough, a good percentage of it at any rate. In those days at Harvard, the study of literature was regarded mainly as a means of studying the history of ideas. What energy was left after the pursuit, capture and analysis of this elusive game was devoted to what was then referred to in hushed tones as "The New Criticism." This new criticism seemed to amount to a word by word dissection of a text with the end of baring its most intimate rhetorical devices, of discovering the pinions of the artist's craft which were presumed to engage with the rack of human sensibilities to produce the exquisite sensations of literature. Ideas stripped of the means of their expression and the color and detail of their provenance, for me at least, are not to conjure with. Thus, in the context of my studies, the grand progress of literary expression, like the triumph of Maximilian, was symbolized by a sort of bare chassis of historical cart laboriously drawn by professors in yoke, occupied by the skeleton of the poet, all sparsely delineated by the new criticism carping in chorus.

However, the past, not only the writing of the past but all of its manifestations, was magical for me. If my fairy godmother had granted me a wish, I should have chosen a time machine in which I could travel to see for myself. But the sort of chewed extract of the past which I was expected to ingest and indeed to excrete on demand was not meaningful to me. In terms of the moment I should have said that it was not relevant. Now, it certainly was relevant to the past. The ideas were there. The rhetorical devices, consciously employed or not, had their stated effects. The ideas were influential, the rhetoric timeless. Thus it was also relevant to the present and future. What I mean to say is that it was not relevant to me. My peculiar combination of talents and sensibilities could not be exercised effectively in that milieu. What I required was an effort of the imagination sufficient to discover a field to which I could usefully contribute and in which my peculiar talent and passion could become involved.

The fact that I turned toward the history of musical instruments and the construction of harpsichords was entirely accidental, and this particular combination of activities could have been replaced by any one of many others. Any successful solution of my dilemma, however, would have had one common denominator which C.V. Wedgewood has called the "romantic approach to history." An irrelevant irony, of course, is that in becoming an historian and exponent of one aspect of the music of the Renaissance and Baroque, usually billed as the natural enemies of the music of the Romantic period, I have done so from motives that would have been entirely comprehensible to Wagner or Schumann.

What I required of any sort of historical discipline was that it both demand and feed an effort of the imagination - that the final result of the perusal of documents or the examination of objects should be the vivid personal sense of the past. I wished to reconstruct and then experience the life of the past, or at least whatever aspects of it were accessible to me in their totality. Miss Wedgewood has described Jacob Burkhardt's purpose in lecturing to his students in Basel as follows: "He said that he wished 'to make every member of my audience feel and know that everyone may and must take independent possession of what appeals to him personally.' 'Take independent possession' is the key phrase, for, ultimately, the understanding of the past, in so far as it is achieved at all, has to be independently achieved by a sustained effort of the imagination working on a personal accumulation of knowledge and experience."

Much of the current opinion that historical studies are irrelevant to the present world and its injustices stems from the essentially romantic and essentially discredited notion that "human life was a constant, forward progress toward an attainable perfection." (Once again I borrow from Miss Wedgewood.)

The past existed chiefly to lead up to, and make way for, the glorious present and still more glorious future . . . . From this it is an easy step to regarding as worthy of study only such institutions and only such persons as can be shown to have some clear connection with the present, and of seeing or imagining in them only such elements as can be made to fit into the splendid story of progress toward the political or social idea as we happen to see it . . . . In spite of these drawbacks, possibly even because of them, the romantic approach to history made the understanding of the past possible in ways never attempted before. The romantics recognized the comprehensive nature of history as a study. 'In the domain of history lies the entire moral world' said Schiller . . . ; for him . . . there was no province of human endeavor outside the scope of history.

And in this definition I would include not only the present but the future. We all live in history. Everything was, is, and will be relevant. However, not only those ideas or techniques or activities which obviously have directly led to an analogue existing at present are of significance. Dead ends, aborted movements, also have their charm and validity. This is nowhere more true than in the history of art and music.

Of course, all of this was not so neatly formulated in my mind during those days in 1946 and 1947. What I felt then was a sort of general malaise, the result of my inability to become personally involved with my studies. I knew that I had to find something more meaningful and I suppose that I defined that merely as something I could do better.

By chance, my stall in the stacks of the Widener Library was on 4 West, the section at that time occupied by books on music, since removed to the Music Library. Often when I should have been analyzing texts according to the currently fashionable modes, I found myself poring over the beautiful picture books devoted to musical instruments. Once I stumbled across a treasure: Heron-Allen's Violin Making As It Was and Is. This young Victorian gentleman's evocation of the world and technique of violin makers, given in some curious way a sweep and perspective by the copious employment of his laboriously acquired Latin tags, fascinated me, and I spent a whole summer making a violin after his instruction. Alas for romantic scholarship. I found his technical devices often deficient and that his varnish wouldn't dry, but the experience was meaningful for all that. The violin still hangs on its nail in a dark closet, a sort of instrument maker's family skeleton. [The violin has since come out of the closet. It is now displayed on a beautiful antique secretary in Diane's and my library. It doesn't suffer at all in comparison with that piece. The quality of workmanship is evident and the varnish is dry. Ed.]

About this time the passionate interest of an old boyhood friend, William Dowd, in the harpsichord began to dimly indicate a path out of the intellectual maze in which I found myself wandering. Together we examined the books in 4 West, attended the concerts of the only practicing harpsichordists in the area, Claude Chiasson and Daniel Pinkham, and gradually concocted the grandiose project of reviving single handed the whole baroque orchestra. I, it was decided, would deal with strings, Dowd with keyboards, and the winds, in some unspecified way, would take care of themselves.

The only makers of such instruments we had ever heard of (at least those not speaking outlandish tongues - and of that I will say more later) were Arnold Dolmetsch in England and John Challis in Detroit. Since my G.I. Bill support was a bit more substantial than that of Dowd, it was decided that I would go to Dolmetsch, Dowd to Challis.

I shall never forget the gracious letter of acceptance I received from Carl Dolmetsch in which he generously waived his usual apprenticeship fee and permitted me to enter his employ as something less than a slave, since I was neither boarded nor clothed as was customary in even the most reactionary of antebellum circles. Several weeks later I found myself walking up the hill toward Beechside, as the house was called, from Haslemere station. Haslemere is set in what an Englishman might call suburbia, but to an American tempered in the crueler manifestations of that drab world, it seemed a fairy village. Cottages, some of them even half-timbered, were picturesquely set in the midst of lawns of an inconceivable greenness. The high road to Guilford up which I carried my box of tools passed along the red brick High Street to become a grassy banked depression between hedges until it emerged on a hillside open to the rolling countryside of Surrey.

Inside Beechside, I found a warren of small workshops in which several men were working who spoke an almost incomprehensible language. First, there was Leslie Ward - tall, lean, bald, shrewd and grasping, with a cockney-timbred voice. He was the works manager and in charge of the harpsichord workshop. George Carley, his brother-in-law, shy and inarticulate, mild and curly-haired, was the head viol maker, assisted by a brassy young woman who I later found spent her lunch hours practicing Max Bruch on the violin. Mr. Ward seriously responded to one remark I made on that incongruity by sighing, "Oh yes, the old man would never have permitted it. We had ideals in those days."

Recorders were turned by a gaggle of other ranks headed by bespectacled Vic Smithers who never ceased wondering that any bird so rare as I could walk the earth. The foreman in the harpsichord shop, by contrast to Smithers, was almost cosmopolitan, upward mobile, too, if the relentless ennoblement of his name meant anything. When I first met him he seemed a pleasant Englishman from the environs of London, called prosaically, Douglas Brown. There must have been some spark of the Celt in him though - he named his daughter something like Gwynneth or Gillian. Soon I noticed that he had become Douglas C. Brown, and not long after it was revealed that this actually stood for Douglas Campbell Brown. The inevitable hyphen eventually cemented his attachment to the proud north. At this stage my connection with Campbell-Brown was severed, but I have since learned that this evolution of Englishman into Scot continued until now one has to deal with an Andrew C. Douglas. I am not sure whether the proper prefix will ultimately be Saint or The.

Brown, as I still think of him, was the neatest workman I have ever known. He arrived each morning in an impeccable pinstripe suit, the paper he had read on the train under his arm. Preparing to work, he would first remove his coat which he carefully hung on a hangar. Then he would pull up the sleeves of his shirt slightly and restrain them over the biceps with rubber bands. No speck of dust, no drop of glue ever seemed to fall on him. At lunch time, he could even work under the car which he occasionally drove to work, emerging with the ball of each finger stained with grease but otherwise unmarked. His bench top was french polished. Never more than one tool at a time rested there. "Lovely work," he used to sarcastically breath as he stooped to examine my amateurish endeavors.

At Dolmetsch's I was permitted to drill identical holes in thousands of small objects, make tea at eleven each morning, and sweep. Occasionally, but not always, my questions were answered. Still, by watching if not by doing, I learned something of wood working and the sort of compulsiveness that makes a craftsman. Of the history of the harpsichord or the glorious examples still extant, I learned nothing.

Curiously enough, in those days, it seemed not have have occurred to anyone that in reviving the instrument in order to bring long neglected music to life, the logical effort would be to follow extant historical examples as models. The earliest modern harpsichords to be made in any quantity were developed in Paris in the last decade of the 19th century by Pleyel, a large piano manufacturer who seems to have turned inward for inspiration, deriving his harpsichord designs as if the project were to develop a plucked piano. Undue emphasis was laid on the ability of the harpsichord to vary its tone color, on the whole not a very central issue. These instruments produced an enormous variety of sounds, all bad. However, Pleyel was blessed by the genius of a young Polish pianist named Wanda Landowska who, in a transcendent exercise of pure imagination, found a way of using even the Pleyel to make viable music. Thus, for nearly fifty years, the movement to revive the literature of the harpsichord was to be dominated by this perversity.

Arnold Dolmetsch, the founder of the firm for which I worked, had been one of the pioneers in the movement to revive old music. Originally a violinist, he had become involved in music for recorders, viols, lute and strings, as well as for the harpsichord. He had formed his family of four children into a touring troupe which performed widely, leaving many traces on the literary as well as the musical life of the time. T.S. Eliot is said to have enjoyed their concerts as a young man. Yates was a personal friend. Bernard Shaw reviewed them favorably. Later, an annual Haslemere festival became traditional for which the faithful regularly journeyed down from London.

In 1908, Dolmetsch was invited by the Chickering Piano Company in Boston to establish a workshop in a corner of their factory to be devoted to the construction of old instruments. Removed somewhat from the proximity of the Pleyel harpsichords, these American harpsichords were the best Dolmetsch ever made. Ralph Kirkpatrick played on Chickerings for years. The American episode in Dometsch's career was followed by a brief interlude in Paris where he directed a harpsichord making project of Pleyel's competitors, Gaveau. But soon he was back in England where he established the workshop I knew and began to turn out instruments of the style which are still made there.

Brown and I used to discuss the lack of correspondence between Dolmetsch harpsichords and old instruments but apparently a combination of ingrained habit in the maker and conditioned expectation in the customer made improvements impossible.

I soon found collections of old harpsichords in London at the Victoria and Albert Museum and in a house owned by the National Trust at Cheyne Walk, a charming tree-lined street on the banks of the Thames. Driven by some sort of uneasy instinct like that of a bird about to build his nest, I began to make notes and measurements of these harpsichords. Brief trips to the continent where I visited the collections in Paris, Brussels, and The Hague added to my archives. I acquired a reader's card for the British Museum and began a desultory search for contemporary descriptions of the instrument.

After about a year at Dometsch's, for various complicated reasons I suddenly found myself without, as they say in England, a place. Now there were not many harpsichord makers anywhere and most in England were terrified of making enemies of the Dolmetschs. The only one I found courageous enough to take me on was Hugh Gough in London. There we formed a two-man shop operating in what had been the front parlor of the dreary row house he occupied in Acton.

Gough, however, was a great improvement over the Dolmetschs. Pert and birdlike, compulsively the correct British gentleman, he was gifted with almost total recall, able to quote at length anything he had ever read. A trifle rakish in his tastes, I found him frightfully sophisticated and was even impelled to make discrete inquiries as to the name of his tailor. Most important, he had seen many old instruments and had notebooks full of details which I devoured. He introduced me to his friend, Donald Boalch, who had been engaged for some years in compiling a dictionary of harpsichord makers. Boalch had also assembled a bibliography of literature on the subject which assisted my efforts in the British Museum considerably.

And so another year passed characterized mainly in my mind by the dank coal smoke smell of London hallways, the cool chill of the air, the round fog-filled dome of the Reading Room at the British Museum, and weekend excursions on the Green Line.

In November of 1949, Dowd and I rented an unheated loft in a ramshackle building on Tremont Street in Boston's South End. We managed to scrape together enough money to buy a circular saw, band saw, drill press, two benches, and a surplus army coal stove which devoured endless quantities of fuel without producing any noticeable heat. Cold winter mornings we huddled around that stove until eleven before we could find courage to venture into the corners of the room. Even so, we did manage to lay down four harpsichords which were epoch making in the simple fact that we were attempting to follow old models. Subsequent opportunities to examine the interior construction of old harpsichords have indicated how far we inadvertently departed from ancient practice, but at least we had been the first to set foot on the new path. That our philosophy of harpsichord making filled a need sensed by many musicians is indicated by the fact that all four were sold before they were finished.

All of our errors, however, were not due to innocent ignorance. In taking that first step on the road to historical rectitude we set foot in a trap that awaits all those who attempt to resurrect, reproduce, restore, or even describe the art and thought of the past. The most servile copyist must make certain decisions, first as to what to copy, then as to what in his model is original and the result of a deliberate act of its maker. We, of course, thought to temper the manifest eccentricities of the past by choosing the best, judiciously adding obvious modern technological improvements and employing certain materials unknown to earlier ages. This was an error which has taken many years to partially overcome. It is a truism that most revivals of antique styles reflect the taste of their own age, not that of the earlier era. Renaissance architecture and ornament is not identical to that of ancient Rome, the school surrounding William Morris was not medieval. Our harpsichords, by their very choice of model, the use of materials such as screws and plywood, music wire and machined brass parts, showed their age and our fatal knowledge of what came after our model in the development of the instrument. Although they were more useful in the performance of old music than the modernist concoctions of Pleyel and Dolmetsch, they testified to our failure to enter in imagination into the world of the past. Although they pretended to follow 17th century Flemish models, they would not have seemed very familiar to Froberger or Chambonniéres.

To make harpsichords is to operate a sort of historical laboratory. One's historical conjectures must be subjected to empirical proof. It is routine enough, even if laborious, to ferret out information about a specific type of old harpsichord. One must measure, examine, weed out later accretions from original parts, compare one extant example to another and then view the resulting data with suspicion in the light of any available written documentation. Then the difficult part comes. The instrument whose design results from these procedures must be appraised, and not on the scale of arbitrary taste but with a taste consciously deformed by acquired knowledge of the expressed opinions and preferences of old authorities and by internal clues left by composers which might indicate their assumptions about the sort of instrument that should be used to realize their compositions. [Italics added. Ed.]

It is plain that this complex process can never become objective. One is always maneuvering between the Scylla of the mindless rationalization of everything for which one can find authority and the Charybdis of an arbitrary subjective judgment. It is hopeless and possibly undesirable to expect to completely eliminate all modern bias in one's judgment and technique. Yet, to the extent that he is able, the harpsichord maker should attempt to direct his decisions by reference to the past and not by an absolute and arbitrary aesthetic standard. The harpsichord is an instrument of the artistic purpose of other times. The syntax of its speech stems from a language that is not ours. To forget these facts is barbarous. [Italics added. Ed.]

Such is what I conceive to be the most effective philosophy for the present day harpsichord maker. Now how does he express it in the products of his trade? His first necessity is to choose a model upon which to exercise his talents and this model must fit certain very specific requirements. First it must be significant - that is to say there must be a body of first class music clearly appropriate to it. The most visible of old harpsichords are not necessarily the most significant. When I was casting about for models, the commonest of large old harpsichords were the English. Every collection seemed to have one and they were seductive. Soundly constructed, of a restful and rational decor, they glowed with the patina of fine old walnut and mahogany. The action had the solidity and precision of an English tall clock. Now there is nothing wrong with attempting to make a harpsichord in the English style, but it is not the place to start. We failed to ask ourselves what music was peculiarly appropriate to the type, that of Dr. Arne, Dr. Pepusch? Even Handel had, after all, been raised on other styles and was rumored to have had a Ruckers.

Before choosing a model, the maker must define his purpose. Is he seeking an instrument for all periods and styles or planning an instrument of more limited purpose? It is easy enough to sneer at general-purpose harpsichords and to insist that they are certain to fail, but how many harpsichords can we expect to place on the platforms of our concert halls? Must the player, like a mouse in a bakery wagon dart from instrument to instrument as his program labors onward in the usual progress from the 16th to 18th century? Therefore, I feel that any commercial maker (and I use the term with a wistful sort of resignation, only wishing it were so) must supply some sort of general purpose instrument at the head of his list. After that he may indulge special clients and his personal pleasure with instruments of more specialized type.

Thus, our first model must not only be historically significant with a large literature for which it is specific, but it must not be entirely inappropriate for as much other literature as possible. This rules out the possibility of a 16th or 17th century model for a general purpose instrument. It would not have enough range nor offer sufficient scope in registration for the later music. This fact also, sadly, condemns the earlier music of the repertoire to less than ideal performance.

In the 18th century, there were four schools of harpsichord making whose products are complex enough to be considered as models for a general purpose harpsichord: the French, the Flemish, The German-Scandinavian, and the English. The last we have already ruled out. A case could be made for each of the others.

The French harpsichords were the vehicle of possibly the most idiomatic of all styles of harpsichord composition. During the 18th century, French artifacts and ideas alike were enormously influential throughout Europe. A fine school of harpsichord making centered in Paris has bequeathed enough instruments to serve as models, and these extant harpsichords are embedded in a matrix of writings, critical and technical, which serve to make the aims and attitudes of makers and players somewhat more accessible than is typically the case. Dowd and I eventually chose the French instruments are our model.

The 18th century Flemish school represented by such makers as Dulcken, Bull and Delin has much to be said for it. To begin with, its style is descended directly from the great 16th and 17th century makers of Antwerp, and yet that style has been sufficiently aggrandized and complicated to make it appropriate to any demand of 18th century music. There is more emphasis on tricky transpositions and a greater length to supply noble basses. The weakness of the Flemish school is an historical one: no composers of real merit can be directly associated with the 18th century Flemish style of harpsichord., Skowroneck has chosen Dulcken as his most usual model.

The German and Scandinavian makers of course had a great school of keyboard composition to lend significance to their efforts. However, their instruments are not as useful models as one might expect. To begin with, the German style was not well unified - it is more difficult to choose a typical example. The large instruments with 16' are each one-of-a-kind. It would be difficult to find enough consensus in design and disposition to settle on a model. They are also rare and it is likely that they always were so. It would seem perverse to choose so unusual an instrument as a model for a general purpose harpsichord. Among the smaller 2 x 8', 1 x 4' German harpsichords one could probably find good models. Few makers have done so.

It is ironic that nearly all 20th century makers are unanimous in declaring the Italian style harpsichord too limited in color and range to stand as a candidate for the general purpose instrument, and yet during the 17th and 18th centuries it more nearly fulfilled that purpose than any other type of harpsichord. Are we once again blinding ourselves to the obvious?

Let us suppose that our fledgling harpsichord maker has finally chosen a model and addresses himself to his task. Exactly which features is he to copy, and which to "improve?" How authentic should his copy be? Must he hand forge his nails, import his wood, and shoot crow for quills?

It used to be said that the essence of a harpsichord was that which could be expressed in a plan view: the lengths of the strings, the point along the string at which it is plucked, the distance from bridge to bentside and to four-foot hitchpin rail, the placement of soundboard ribs, the thickness of the soundboard, and, of course, the range and disposition of the instrument. Many "copies" have been made which do not resemble their prototypes except in having a common plan.

As the movement toward authentic style and technique in performance becomes more refined, the actions of harpsichords have come under closer scrutiny. Now we are concerning ourselves more with the details which can be imparted by the elevation: the lengths, thicknesses, weighting, balance points and method of guiding the key levers, the key dip and point at which the dip is arrested, the weight of the jack, the type of cloth padding at various points in the action.

These things are all essential, but even attention to this formidable list seems insufficient. There is new emphasis on materials. Differences which seemed insignificant some years ago now appear vital. Delrin quill replaced leather plectra because it sounded more like crow quill. Now in subtle insight, several makers have discovered that crow quill sounds even more like crow quill than Delrin does. The difference is small. Note for note they are indistinguishable, but a whole instrument voiced in quill is not identical to an instrument voiced in Delrin.

Old music wire was softer and more flexible than modern high tensile steel wire. The tone of a harpsichord string in this wire is more transparent and interesting than the plain and clear sound of an instrument strung with modern wire. This is particularly true of the harpsichord tuned to a lower "old" pitch.

American makers have been satisfied to make soundboards of Sitka spruce for years. Now, several makers are importing Norway spruce, the species most commonly used in old harpsichords.

Case materials have become more esoteric. A few years ago, most modern harpsichords were made of plywood or stable woods such as mahogany. Now makers are beginning to use the lime or poplar most often found in continental harpsichords.

They endeavor to duplicate the light touch, the silvery , vibrant tone, light yet expressive, which is found in the best old harpsichords. The most successful makers have achieved this end by attention to detail and by applying a mixture of scientific common sense, historical imagination, and aesthetic perception to the problem. Of course, one man's common sense is another man's gobbledygook. Only the result can finally resolve the argument.

One pitfall is apparent. Just as the amateur cook is likely to believe two measures of an ingredient to be more effective than one, many makers in the endeavor to be super authentic have actually abandoned all historical verisimilitude in indulging their personal caprice. The most common eccentricity of this sort is the pursuit of extreme lightness of case weight, either by choice of material or by the thinness of parts. Cedar, a very light wood, has been used by several makers in reproductions of North European harpsichords. Old instruments of this type were never made of cedar, nor were their dimensions ordinarily reduced to the values one now sometimes finds.

Another exaggeration of this kind is what at first would seem a good thing: over-fastidious craftsmanship. Old harpsichords by good makers like Blanchet, Dulcken, or Hass were neatly and soundly made but they were not fussy. In morality it is much easier to be correct than just, and in workmanship nothing is more difficult to recapture than that sort of secure and rapid expression given their concepts by old makers. It is easy, if expensive, to agonize over a joint or finish. It is difficult to carry it out neatly and adequately the first time. Competence is the ability to perform a task well and quickly. Overly slow work leaves characteristic flaws as plain to the connoisseur as the errors of the careless workman. [Italics added. Ed.]

We have now stood at our young harpsichord maker's shoulder ever ready with prolix advice as he has chosen his model, decided which parts of it to copy, and determined the materials he is to employ. Let us leave him now to turn out several instruments in its image and rejoin him only as he considers what steps she should take next.

He will have noticed what seem to him the beauties of his model and its weaknesses. He must assess the latter with particular care, constantly on guard against applying inappropriate criteria from the modern age. However, by comparing his instrument to antiques and good copies of other models, he should become aware of certain shortcomings. He is now ready to move on to the next stage which is to reproduce not a certain instrument, but a style of instrument - to choose the best from several antiques and yet make a harpsichord which is recognizable as stemming from a certain school. This process obviously requires a more complete control of his material on the part of the maker than the process of making a copy. Now he must decide not only what was done but what was central to the concept. If there are two means of achieving an objective, he must recognize that fact and not attempt both ways simultaneously. Instruments made as representations of a school are certainly more risky to attempt. Fewer will be relevant to the music. They will be more likely to fall out of fashion as they age. Yet, I suppose that the rewards of success are greater.