THEHISTORY OF MUSIC LIBRARY |
SECTION VI The English School It is somewhat remarkable
that the Continental writers on the Violin should have omitted to
mention any English maker, either ancient or modern. Such an omission
must have occurred either from want of information concerning our
best makers, or, if known, they must have been deemed unworthy of
the notice of our foreign friends. There is no mention of an English
maker in the work of F�tis, "Antoine Stradivari," 1856,
although numerous very inferior German and Italian makers are quoted.
The same omission is also conspicuous in "Luthomonographie
Historique et Raisonn�," 1856, and Otto's "Ueber den Bau
der Bogeninstrumente," &c., 1828. It may be that Continental
connoisseurs have credited themselves with the works of our best
makers, and expatriated them, while they have inexorably allowed
bad English Fiddles to retain their nationality. However, it is
my desire that my foreign brothers should be enlightened on this
point, and in all candour informed of the array of makers that England has at different times
produced, and is yet capable of producing, did but the new Violin
command the price that would be a fair return for the time and skill
required in the production of an instrument at once useful and artistic.
It will be my endeavour to show forth the qualities of those of
our makers whose names, as yet, seem never to have crossed the Channel,
so that when these pages on the English School are read by distant
connoisseurs, and the merits and shortcomings of the makers therein
are fairly weighed by them, the good shall be found so to outweigh
the indifferent as to entirely change the opinions formed of us
as makers of the leading instrument.
Until the early nineteenth
century makers of Violins in England would appear to have been comparatively
numerous, if we take into consideration the undeveloped state of
stringed instrument music at that period in this country. Among
those makers were men of no ordinary genius—men who worked
lovingly, guided by motives distinct from commercial gain, so long
as they were allowed to live by their work. When, however, the duties
on foreign musical instruments were removed, the effect was to partially
swamp the gallant little band of Fiddle-makers, who were quite unable
to compete with the French and German makers in price (not excellence,
be it distinctly understood, for we were undoubtedly ahead of our
foreign competitors, both in style and finish,
at this period). The prices commanded by many English makers previous
to the repeal of the duty were thoroughly remunerative. Five to
twenty pounds were given for English Violins, while Violoncellos
and Tenors commanded prices proportionately high. The English Violin-makers
were thus enabled to bestow artistic care in the making of their
instruments. When, however, they were suddenly called upon to compete
on equal terms with a legion of foreign manufacturers, the result
was not so much that their ardour was damped, as that they themselves
were extinguished, and served as another instance of the truth of
the adage that "the good of the many is the bane of the few." In matters of magnitude,
whether artistic or otherwise, competition is undoubtedly healthy,
there being always a small body of patrons who are willing to check
the tendency to deteriorate, common to all productions, by encouraging
the worker with extra remuneration, in order that a high degree
of excellence may be maintained; but in matters confined to a small
circle, as in the case of Violin-making, the number of those willing
to encourage artistic workmanship is so minute as to fail even to
support one maker of excellence, and thus, when
deprived suddenly of its legitimate protection, the art, with other
similar handicrafts, must drift into decadence. If we look around
the Violin world, it is everywhere much the same. In Italy there
is no Stradivari in embryo, in France no coming Lupot, in Germany
no Jacob Stainer, and in England no future Banks or Forster. Why
so? The answer is twofold. Partly there is fault in the demand,
arising from the marked preference of this age for cheapness at
the expense of goodness; partly, too, there is a fault in the supply,
a foolish desire on the part of the makers to give maturity to their
instruments, wherein they always completely fail, yet they will
not give up their conceit. Here, again, were we dealing with matters
of greater magnitude, the evil influence would be lessened, the
artistic impulses would still be felt, though in a less degree;
whereas so contracted is the circle of the Violin world, that under
any stress the support given to makers willing to bestow an artist's
care on their work is totally inadequate. The case of modern
Violin-makers is unfortunate. Old Violins being immeasurably superior
to modern productions, the demand must necessarily set steadily
for the former, and the modern maker has only the few patrons of
new work to support him. It cannot be expected that the players
of to-day should patronise the modern Violin in order that the next
generation should reap the benefit. Years since it was quite a different
matter. The makers were well paid for their work, and new instruments
were then made to supply wants similar to those which the horrid
Mirecourt or Saxon copy fulfils at present. As with other things,
so is it also with Violins; if they are to be produced with the
stamp of artistic merit, they must be paid for accordingly; without
patronage the worker necessarily becomes careless. Finding that
his skill fails to attract attention, he gradually sinks down into
the mere routine of the ordinary workman. When Italy shone brightest
in art, the patronage and remuneration which the workers received
was considerable. Had it been otherwise, the powers of its Raffaele,
its Cellini, and last (though not least to the admirers of the Violin),
its Stradivari, would have remained simply dormant. Art, like commerce,
is regulated in a great measure by supply and demand. In Raffaele's
day, sacred subjects were in demand; the Church was his great patron,
and aided him in bringing forth the gift which nature had implanted
within him. In modern times, landscape-painting became the favoured
subject, particularly in England; the result of which preference
has been to place us in the foremost rank in that branch of art.
The stage furnishes another instance of the effect that patronage
has in bringing forth latent talent. If the history of dramatic
art be traced, it will be found that its chief works were written
when the taste of an appreciative public could be securely counted
upon. As it waned, so the writers of merit became rarer; or perhaps
it would be more correct to say, the plays produced became less
meritorious, the authors being constrained to pander to the prevailing
tastes. As further evidence
of the effect of patronage on art, a case in point is found in the
manufacture of Venetian glass. The Venetians, centuries ago, became
famous for their works in glass, and the patronage they enjoyed
was world-wide; but their country being thrown into an unsettled
condition, capital drifted from it, until the blowing of glass,
together with other industries, was comparatively extinguished.
Within recent years the art of making glass has shown signs, even
in Venice itself, of reviving with all its former vigour in the
workshops of Salviati, the success of which is due in great measure
to English capital. With regard to English
Violin manufacture, there would be no reason why Violins should
not, at the present moment, be produced in England which should
fully reach the standard of merit maintained in our forefathers'
days, if only the patronage of the art occupied a larger area. The
present dearth of English makers does not arise from any national
want of talent for this particular handicraft; in fact, we have
plenty of men quite as enthusiastic as our foreign friends for a
vocation which, in England also, must be pronounced to be alike
venerable in its antiquity and famed for the dexterity of its genius. The earliest makers
of Viols in England seem to have been Jay, Smith, Bolles, Ross,
Addison, and Shaw, names thoroughly British. We may take this as
good evidence that the making of Viols in England originated with
the English, and was not commenced by settlers from the Continent.
Doubtless the form of the English Viol and its brethren was taken
from the Brescian makers, there being much affinity between these
classes of instruments. In the few Violins extant by Christopher
Wise the Italian character is very striking. In them we see a flat
model, excellent outline, and varnish of good quality. The Viols
of Jay have the same Italian character. Later on we have names of
some reputation—Rayman, Urquhart, and Barak Norman. In the
absence of any direct evidence as regards the nationality of these
makers it is requisite to endeavour to trace the style belonging
to their works. It will be observed that there was a great improvement
in the style of work and varnish of instruments made in England,
commencing with the time of Rayman, and it is probable that this
step in advance was obtained from intercourse with Italy or the
German Tyrol. Starting with Rayman, there is a German ring in the
name which makes me think that he came from Germany, and, if so,
brought with him the semi-Italian character of work common to the
makers who lived so near Brescia. If the work and style of Rayman
be carefully examined, it will be seen that there is much in common
with the inferior Brescian makers. The outline is rugged, the sound-hole
is of that Gothic form peculiar to Brescia; the head is distinct
from that of the early English type. At the same period Urquhart
made instruments of great merit, the varnish of which is superior
to that of Rayman, but is evidently composed of similar ingredients.
Its superiority may have arisen from a different mode of mixing
only. The name of Urquhart has a North British sound, and it is
probable that he was born in Scotland, and settled in London as
an assistant to Rayman, who would impart to him the style of foreign
work. The semi-Italian
character pervading the instruments made in England at this period
seems to have culminated in the productions of Barak Norman, whose
best works bear even a more marked Brescian character than those
of Rayman. The model varies very much, sometimes being high, at
other times very flat; in the latter case the results are instruments
of the Maggini type. Barak Norman frequently double-purfled his
instruments, and inserted a device in the purfling, evidently following
Maggini in these particulars. With Barak Norman ends the list of
English copyists of the Brescian makers. We now arrive at
the copyists of Jacob Stainer and the Amati, a class of makers who
possessed great abilities, and knew how to use them. The first name
to be mentioned is Benjamin Banks, of Salisbury, who may with propriety
be termed the English Amati. He was the first English
maker who recognised the superior form of Amati's model over that
of Stainer, and devoted all his energies to successful imitation.
Too much praise cannot be lavished on Banks for the example which
he selected for himself and his fellow-makers. Next follow the names
of Forster, Duke, Hill, Wamsley, Betts, Gilkes, Hart, and Kennedy,
together with those of Panormo, Fendt, and Lott, who, although not
born in England, passed the greater part of their lives here, and
therefore require to be classed with the English School. The mention
of these makers will bring the reader to the present time. Upon scanning this
goodly list, there will be found ample evidence that we in England
have had makers of sufficient merit to entitle us to rank as a distinct
school—a school of no mean order. We may therefore assume
that the Continental writers who from time to time have published
lists of makers of the Violin, and have invariably ignored England,
have erred through want of information regarding the capabilities
of our makers, both ancient and modern. The following list
will be found to enumerate nearly the whole of the English makers,
and indicate the distinctive character of their respective works. English Makers ABSAM, Thomas, Wakefield,
1833. ADAMS, Garmouth,
Scotland, 1800. ADDISON, William,
London, 1670. AIRETON, Edmund.
Was originally employed in the workshop of Peter Wamsley, at the
"Harp and Hautboy," in Piccadilly. He made a great many
excellent Violins and Violoncellos, and chiefly copied Amati. Varnish
of fair quality; colour yellow. He died at the advanced age of 80,
in the year 1807. ALDRED, ——,
about 1560. Maker of Viols. ASKEY, Samuel, London,
about 1825. BAINES, about 1780. BAKER, ——,
Oxford. Mention is made of a Viol of this maker in the catalogue
of the music and instruments of Tom Britton, the small-coal man. BALLANTINE, Edinburgh
and Glasgow, 1850. BANKS, Benjamin,
Salisbury, born 1727, died 1795. To this famous maker must be given
the foremost place in the English School. He was a thorough artist,
and would not have been thought lightly of had he worked in Cremona's
school, and been judged by its standard. This may be considered
excessive praise of our native maker; but an unprejudiced judge
of work need only turn to the best specimens of Banks's instruments,
and he will confess that I have merely recorded a fact. Banks is, again,
one of the many instances of men who have gained a lasting reputation,
but whose histories have never reached the light to which their
names have attained. How interesting would it be to obtain the name
of his master in the knowledge of making instruments! No clue whatever
remains by which we could arrive at a satisfactory conclusion on
this point. That he was an enthusiast in his art is certain, and
also that he was aware to some extent that he possessed talent of
no mean description. This is evidenced by the fact that many of
his instruments are branded with the letters B. B. in several places,
as though he felt that sooner or later his works would be highly
esteemed, and would survive base imitations, and that by carefully
branding them he might prevent any doubt as to their author. Many
of his best instruments are found to have no brand: it would seem,
therefore, that he did not so mark them for some time. He appears
to have early shown a preference for the model of Niccol� Amati,
and laboured unceasingly in imitation of him, until he copied him
with an exactness difficult to surpass. Now that time has mellowed
his best works, they might pass as original Amatis with those not
perfectly versed in the characteristics of the latter. Many German
makers excelled as copyists of Amati; but these makers chiefly failed
in their varnish, whereas Banks was most happy in this particular,
both as regards colour and quality. If his varnish be closely examined,
its purity and richness of colour are readily seen. It has all the
characteristics of fine Italian varnish, being beautifully transparent,
mellow, and rich in its varieties of tints. It must be distinctly
understood that these remarks apply only to the very finest works
of this maker, there being many specimens which bear the label of
Banks in the framing of which he probably took but a small share,
leaving the chief part to be done by his son and others. Banks cannot
be considered as having been successful in the use of his varnish
on the bellies of his instruments, as he has allowed it to clog
the fibre, a blemish which affects the appearance very much, and
has been the means of casting discredit on the varnish among those
unacquainted with the real cause. The modelling is executed with
skill. Fortunately, sufficient wood has been left in his instruments
to enable time to exert its beneficial effects, a desideratum overlooked
by many makers of good repute. The only feature of his work which
can be considered as wanting in merit is the scroll, which is somewhat
cramped, and fails to convey the meaning intended, viz., the following
of Amati; but as this is a point having reference to appearance,
and therefore solely affecting the connoisseur, it may be passed
over lightly, and the more so when we consider that Banks was not
the only clever workman who has failed in head-cutting. He made
Violins, Tenors, and Violoncellos, all excellent; but the last-named
have the preference. His large Violoncellos are the best; those
of the smaller pattern are equally well made, but lack depth of
tone. The red-varnished instruments are the favourites. BANKS, Benjamin,
son of the above, born in September, 1754; died January, 1820. Worked
many years with his father at Salisbury, afterwards removed to London,
and lived at 30, Sherrard Street, Golden Square. BANKS, James. Brother
of the above. For some years carried on the business of his father
at Salisbury, in conjunction with his brother Henry. They ultimately
sold the business and removed to Liverpool. The instruments of James
and Henry Banks are of average merit. BARNES, Robert, 1710.
Worked with Thomas Smith at the "Harp and Hautboy" in
Piccadilly. Afterwards partner with John Norris. BARRETT, John, 1714.
An average workman, who followed the model of Stainer. His shop
bore the sign of the "Harp and Crown." Barrett was one
of the earliest copyists of Stainer, and in the chain of English
makers is linked with Barak Norman and Nathaniel Cross. The wood
is generally of a very good quality, the varnish yellow. BARTON, George, Old
Bailey, London, about 1780-1810. BETTS, John, born
1755, at Stamford, Lincolnshire, died in 1823. Became a pupil of
Richard Duke. He commenced business in one of the shops of the Royal
Exchange, where he soon enjoyed considerable patronage. John Betts
does not appear to have made a great number of instruments, but
employed many workmen, into whose instruments he inserted his trade
label. He was, perhaps, the earliest London dealer in Italian instruments.
His quaintly-worded business card runs:— "John Betts,
Real Musical Instrument Maker, at the Violin and German Flute, No.
2, under the North Piazza of the Royal Exchange, makes in the neatest
manner, Violins the patterns of Antonius Stradivarius, Hi�ronymus
Amati, Jacobus Stainer, and Tyrols. Equal for the fine, full, mellow
tone to those made in Cremona. Tenors, Violoncellos, Pentachords,
&c., &c., &c." The sound-holes of
Betts' instruments are rather wide; broad purfling; scroll well
cut. BETTS, Edward, nephew
of John Betts; was a pupil of Richard Duke, whose work he copied
with considerable skill. Of course, in trying to imitate Duke he
was copying Amati, Richard Duke having spent his life in working
after the Amati pattern, without attempting to model for himself.
The care bestowed by Edward Betts on his instruments was of no ordinary
kind. The workmanship throughout is of the most delicate description;
indeed, it may be said that neatness is gained at the expense of
individuality in many of his works. Each part is faultless in finish,
but when viewed as a whole the result is too mechanical, giving
as it does the notion of its having been turned out of a mould.
Nevertheless, this maker takes rank with the foremost of the English
copyists, and in his instruments we have as good specimens of undisguised
work as can be readily found. They will be yearly more valued.
BOLLES, ——,
An early maker of Lutes and Viols. BOOTH, William, 1779
to about 1858, Leeds. BOOTH, ——,
son of the above, Leeds, died 1856. BOUCHER, ——,
London, 1764. BROWN, James, London,
born 1770, died 1834. Worked with Thomas Kennedy. BROWN, James, London,
son of the above, born 1786, died 1860. BROWNE, John, London,
about 1743. Worked at the sign of the "Black Lion," Cornhill.
Good work. Amati pattern. Scroll well cut; hard varnish. CAHUSAC, ——,
London, 1788. Associated with the sons of Banks. CARTER, John, London,
1789, worked with John Betts, and afterwards at Drury Lane on his
own account. CHALLONER, Thomas,
London. Similar to Wamsley. COLE, Thomas, London,
1690. COLE, James, Manchester,
19th century. COLLIER, Samuel,
1750. COLLIER, Thomas,
1775. COLLINGWOOD, Joseph,
London, 1760. CONWAY, William,
1750. CORSBY, ——,
Northampton, 1780. Chiefly made Double-Basses. CORSBY, George. Lived
upwards of half a century in Princes Street, Leicester Square, where
he worked and dealt in old instruments. CRAMOND, Charles,
Aberdeen, 19th century. CRASK, George, Manchester.
He made a large number of instruments, chiefly imitations. CROSS, Nathaniel,
London, about 1700-50. Worked with Barak Norman. He made several
good Violins. Purfling narrow; excellent scroll. CROWTHER, John, 1760-1810. CUTHBERT, London,
17th century. Maker of Viols and Violins. Many of the latter have
merit. Model flat, and wood of good quality. Very dark varnish. DAVIDSON, Hay, Huntley,
1870. DAVIS, Richard. Worked
with Norris and Barnes. DAVIS, William, London.
Succeeded Richard Davis in the business now carried on by Edward
Withers. DEARLOVE, Mark, Leeds. DELANY, John, Dublin.
Used two kinds of labels, one of them very small— DENNIS, Jesse, London,
1805. DEVEREUX, John, Melbourne.
When in England he worked with B. Simon Fendt. DICKINSON, Edward,
London, 1750. Made instruments of average merit. The model is high. DICKESON, John, 1750-80,
a native of Stirling. He would seem to have lived in various places,
some instruments dating from London and some from Cambridge. He
was an excellent workman, and chiefly copied Amati. His work much
resembles that of Cappa. DITTON, London, about
1700. Mention is made of an instrument by this maker in Tom Britton's
Catalogue. DODD, Thomas, son
of Edward Dodd, of Sheffield. He was not a maker of Violins. Numerous
instruments bear his name, but they are the work of John Lott and
Bernard Fendt. The merit of these instruments is of the highest
order, and they are justly appreciated by both player and connoisseur.
Thomas Dodd deserves to be mentioned in terms of high praise, notwithstanding
that the work was not executed by him, for his judgment was brought
to bear upon the manufacture during its various stages, and more
particularly in the varnishing, in which he took the liveliest interest.
He had a method of mixing colours, the superior qualities of which
he seems to have fully known, if we may judge from the note on his
labels, which runs thus: "The only possessor of the recipe
for preparing the original Cremona varnish. Instruments improved
and repaired." This undoubtedly savours of presumption, and
is certainly wide of the truth. Nevertheless there is ample evidence
that the varnish used by Thomas Dodd was very excellent, and had
a rich appearance rarely to be met with in instruments of the English
school. Dodd was encouraged in the art of varnish-making by persons
of taste, who readily admitted the superior qualities of his composition,
and paid him a handsome price for his instruments. He was thus enabled
to gratify his taste in his productions by sparing no means to improve
them. He ultimately attained such a reputation for his instruments
as to command no less a sum than �40 or �50 for a Violoncello. Commanding
such prices, it is evident that he spared no expense, or, what was
to him a matter of still greater importance, no time. He was most
particular in receiving the instruments in that incomplete stage
known in the trade as "in the white," i.e.,
without varnish. He would then carefully varnish them with his own
hands, guarding most warily the treasured secret of the composition
of his varnish. That he never departed from this practice may be
inferred from the fact that the varnish made by the workmen in his
employ, apart from the establishment, for their own instruments,
is of an entirely different stamp, and evidently shows that they
were not in their master's secrets. The instruments bearing
the Dodd label are not valued to the extent of their deserts, and
there can be but little doubt that in the course of time they will
be valued according to their true merits. They were made by men
of exceptional talent, who were neither restricted in price nor
material. Under such favourable conditions the results could not
fail to be good. DODD, Thomas, London.
Son of Thomas Dodd, musical instrument dealer, of St. Martin's Lane.
The father, although not a maker of Violins, possessed excellent
judgment, both as regards work and makers, which enabled his son
to profit considerably during his early years whilst working with
Fendt and Lott. DORANT, William,
London, 1814. DUKE, Richard, worked
from 1750-80. The name of this maker has long been a household word
with English Violinists both amateur and professional. Who has not
got a friend who is the fortunate owner of a veritable "Duke"?
The fame of His Majesty Antonio Stradivari himself is not greater
than that of Richard Duke in the eyes of many a Fiddle fancier.
From his earliest fiddling days the name of Duke became familiar
to him; he has heard more of him than of Stradivari, whom he somehow
confuses with Cremona. He fondly imagines that Cremona was a celebrated maker,
and Stradivari something else; inquires, and becomes more confused,
and returns again to "Duke," with whom he is thoroughly
at home. Many excellent judges
have wondered how it came to pass that Richard Duke should have
been so highly valued, there being, in their estimation, so little
amongst his remains worthy of the reputation he gained. The truth
is that no maker, with the exception of the great Cremonese artists,
has been so persistently counterfeited. The name of Duke has been
stamped upon every wretched nondescript, until judges who had not
the opportunity of seeing the genuine article mistook the copies
for the original, and hence the confusion. When, however, a really
fine specimen of Duke is once seen, it is not likely to be forgotten.
As copies of Amati such instruments are scarcely surpassed, varnish,
work, and material being of the best description. The copies of
Stainer were not so successful. DUKE, Richard, London.
Son of the above. DUNCAN, ——,
Aberdeen, 1762. DUNCAN, George, Glasgow,
contemporary. EGLINGTON, ——,
London, 1800. EVANS, Richard, London,
1750. His label is a curiosity— FENDT, Bernard, born
at Innsbruck, in the Tyrol, 1756, died 1832. He was evidently a
born Fiddle-maker, genius being stamped, in a greater or less degree,
upon all his works. To Thomas Dodd belongs the credit of bringing
his talent into play. Dodd obtained the services of Fendt upon his
arrival in England, which the latter reached at an early age. He
remained with Dodd many years, frequently making instruments with
John Frederick Lott. The instruments so made bear the label of Thomas
Dodd. Lott being also a German, reciprocity of feeling sprung up
between him and Fendt, which induced Lott to exchange the business
to which he was brought up for that which his fellow countryman
Fendt had adopted, and henceforth to make Violins instead of cabinets.
By securing the services of these admirable workmen, Dodd reaped
a rich harvest. He found in them men capable of carrying out his
instructions with an exactness that could not be surpassed. Dodd
was unable to use the tools himself; but in Fendt and Lott he had
men who were consummate masters of them. When the instruments were
finished, as far as construction was concerned, they were clothed
in coats of the master's livery—"Dodd's varnish,"
the secret of making which he kept carefully to himself. With these
coats of varnish upon them the work was doubly effective, and every
point of excellence was made to shine with the happiest effect.
Upon leaving the workshop of Thomas Dodd, Bernard Fendt worked for
John Betts, making many of those copies of Amati which are associated
with the name of Betts, and which have so high a value. Although Fendt was
German by birth, his style of work cannot be considered as German
in character. Having early quitted his post of trade in Paris for
England, and having in this country placed himself under the guidance
of Dodd, who steadfastly kept before his workmen the originals of
the great Italian masters for models, his work acquired a distinctive
stamp of its own, and in its turn gave rise to a new and independent
class of makers. FENDT, Bernard Simon,
London, born in 1800, died in 1852. Son of the above. He was an
excellent workman. It is to be regretted that he did not follow
the excellent example set by his father, and let time do its work,
without interruption, upon his instruments. Had he done so they
would, in many instances, have been equal to those of his parent;
but, unfortunately, he worked when the mania for obtaining supposed
maturity by artificial means was at its height, and shared the general
infatuation, and, in consequence, very frequently destroyed all
the stamina of his instruments. Subsequently he became a partner
of George Purdy, and carried on a joint business at Finch Lane,
in the City of London, from whence most of his best instruments
date. Purdy and Fendt had also a shop in the West End about 1843.
He was a most assiduous worker. The number of Violins, Tenors, Violoncellos,
and Double-Basses that he made was very great; indeed, his reputation
would have been greater had he been content to have made fewer instruments
and to have exercised more general care. His copies of Guarneri
are most numerous, numbering some hundreds. They are mostly varnished
with a glaring red colour, of a hard nature. He made many good Double-Basses
of the Gasparo da Sal� form, the varnish on which is superior to
that on his Violins. He made also an excellent quartette of instruments—Violin,
Viola, Violoncello, and Double-Bass—for the Exhibition of
1851. They were certainly the best contemporary instruments exhibited,
but he failed to obtain the prize medal. FENDT, Martin, London,
born 1812, died 1845. Brother of the above. Worked for Betts. FENDT, Jacob, London,
born 1815, died 1849. Third son of Bernard Fendt. The best maker
among the sons of Bernard. His instruments are beautifully finished,
and free from the stereotyped character belonging to those of his
brother Bernard. As specimens of the imitator's art they are unsurpassed.
One cannot but regret that such a consummate workman should have
been obliged to waste his energies in making new work resemble that
of a hundred years before. The patronage that he obtained was not
of much value, but had he brought his work into the market in its
natural condition he could not have lived by his trade. He was,
therefore, compelled to foster that which he no doubt felt to be
degrading. The copies of Stradivari by Jacob Fendt are among his
best efforts. The work is well done; the discoloration of the wood
cleverly managed, the effects of wear counterfeited with greater
skill than had ever been done before, and finally, an amount of
style is thrown into the work which transcends the ingenuity of
any other copyist. Had he been allowed to copy the form of the old
masters, as Lupot did, without imitating the actual wear of the
instrument, we should have had a valuable addition to our present
stock of instruments of the Panormo class. FENDT, Francis, London.
Fourth son of Bernard; also worked in Liverpool about 1856. FENDT, William, London,
born 1833, died 1852. Son of Bernard Simon Fendt. Was an excellent
workman, and assisted his father in the manufacture of several of
his Double-Basses. FERGUSON, Donald,
Huntley, Aberdeenshire. FIRTH, G., Leeds,
1836. FORSTER, William,
born in 1713, died 1801. The family of the Forsters have played
no unimportant part in the history of Violins. The attention they
commanded as makers, both from artists and amateurs, has probably
never been equalled in England. Their instruments claimed attention
from the moment they left their makers' hands, their construction
being excellent in every way. William Forster was a native of Brampton,
in Cumberland, where he followed the trade of a spinning-wheel maker,
occupying his spare time in the making and repairing of Violins
and musical instruments generally. His labours, as far as they relate
to Violin-making, appear to have been of a very unpretending nature,
but they served to impart a taste for the art to his son William,
who was the best maker of the family. FORSTER, William,
London, born 1739, died 1807. Son of William Forster mentioned above.
Worked with his father at Brampton in Cumberland, making spinning-wheels
and Violins—two singularly diverse occupations. It was, however,
to the latter industry he gave the most attention, and he soon became
the great maker of the neighbourhood. He afterwards added another
string to his bow, viz., that of playing country-dances at the village
festivities. Thus armed with three occupations, he must have been
well employed. He seems to have early discovered that his abilities
required a larger field in which to show themselves to advantage,
and accordingly took the usual course in such circumstances—came
to the Metropolis, in which he settled about the year 1759. He soon
obtained employment at a musical instrument seller's on Tower Hill,
and gave up, then and for ever, the making of spinning-wheels, while
by throwing all his soul into the manufacture of Violins he soon
gave his master's patrons the highest satisfaction. He ultimately
commenced business on his own behalf in the neighbourhood of Duke's
Court, St. Martin's Lane, where his abilities attracted considerable
attention, and secured him the patronage of the dilettanti in
the musical world. For several years he followed the path trodden
by the makers of the period, and copied Stainer. His instruments
of this date are very excellent both in workmanship and material,
but are not equal to those of the Amati pattern, which he commenced
to make about the year 1770. These are beautiful works, and have
a great charm from their being so varied. Some are copies of Antonio
and Girolamo Amati, variously modelled; others are copies of Niccol�
Amati. The wood and varnish also vary very much, but the high standard
of goodness is well maintained throughout. His varnish was, during
the last twenty years of his life, very fine in quality, and in
the manufacture of it he is said to have been assisted by a friend
who was an excellent chemist. He made only four Double-Basses, three
of which were executed for the private band of George III. Forster's
instruments were the favourite equipment of Robert Lindley, and
their value in his day was relatively far higher than at the present
moment. When Lindley died attention was turned to Italian Violoncellos,
and a vast number having been brought to England, the value of Forster's
productions was very considerably depreciated; now, however, that
the cultivation of stringed instrument music has been so much extended,
they are rapidly rising again to their former level, Italian instruments
being a luxury not obtainable by every one, and age having so benefited
the tone of Forster's Violoncellos as to render them excellent substitutes. FORSTER, William,
London, born in 1764, died 1824. Son of William Forster, the second
of the family. Although this maker did not attain to the celebrity
of his father, his instruments are often fully as good. The workmanship
is very neat, and the modelling excellent, the varnish being equal
to that on his father's instruments. FORSTER, William,
London, born in 1788, died 1824. Son of William Forster, mentioned
above. He was a very good workman: he made but few instruments. FORSTER, Simon Andrew,
London, born in 1801, died about 1870. Brother of William, mentioned
above. He learned his business from his father and Samuel Gilkes,
who worked for William Forster. He made several instruments between
the years 1828 and 1840, which are of average merit. Best known
as joint author with W. Sandys of a "History of the Violin"
(London, 1864). FRANKLAND, ——,
London, about 1785. FURBER, ——,
London. There were several makers of this family, some of whom worked
for Betts, of the Royal Exchange. Many of their instruments are
excellent, and should unquestionably be more valued than they are.
John Furber made several Violins of the grand Amati pattern, and
also copied with much ability the "Betts" Stradivari,
when the instrument belonged to Messrs. Betts in the Royal Exchange,
for whom he worked. FURBER, Henry John,
son of John Furber, London. He has made several excellent instruments,
and maintained the character for good workmanship which has been
associated with the name of Furber for upwards of a century. GIBBS, James, 1800-45.
Worked for Samuel Gilkes and others. GILKES, Samuel, London,
born in 1787, died in 1827. Was born at Morton Pinkney, in Northamptonshire.
He became an apprentice of Charles Harris, whose style he followed
to some extent. Upon leaving Harris he engaged himself to William
Forster, making many instruments for him, retaining, however, all
the features of the style of Harris. In the year 1810 he left the
workshop of Forster, and commenced business on his own account in
James Street, Buckingham Gate, where the few instruments bearing
his name were made. Too much cannot be said in praise of much of
the work of this excellent maker. The exquisite finish of many of
his instruments evidences that the making of them was to him a labour
of love. Amati was his favourite model. GILKES, William,
London, born 1811, died 1875. Son of Samuel Gilkes. Has made a great
number of instruments of various patterns, chiefly Double-Basses. GOUGH, Walter. An
indifferent workman. HARBOUR, ——,
London, about 1785. HARDIE, Matthew,
Edinburgh, date from about 1800. He was the best maker Scotland
has had. The model is that of Amati; the work throughout excellent.
The linings are mostly of cedar. He died about 1825-26. HARDIE, Thomas, Edinburgh.
Worked with his father, Matthew Hardie. He was born in 1804, died
1856. HARE, John, London.
About 1700. His label shows that he was in partnership, his name
being joined to that of Freeman, and the address is given as "Near
the Royal Exchange, Cornhill, London." Much resembles the work
and style of Urquhart. Varnish of fine quality. HARE, Joseph, London,
probably a son of John Hare, above-mentioned. Varnish of excellent
quality. HARRIS, Charles,
London, 1800. This maker is known only to a few dealers, as he made
chiefly for the wholesale merchants of his day. His name was rarely
affixed to his instruments, but those thoroughly acquainted with
his work agree in giving him a foremost place among the makers of
this country. He was, like many other makers of that period, engaged
in two occupations differing very much from each other, being at
the same time a Custom-house officer and a maker of Violins. The
former circumstance brought him into contact with mercantile men,
and enabled him to obtain commissions to make Violins for the export
trade. His business in this direction so increased that he obtained
the services of his relative, Samuel Gilkes, as his assistant. He
never aimed at producing a counterpart of the instrument that he
copied by resorting to the use of deleterious means to indicate
upon the surface of an instrument the ravages of time. He faithfully
copied the form, and thus did what Lupot was doing at the same period.
The finish of these instruments is excellent, and as they are covered
with a good quality of varnish, they have every recommendation of
appearance. HARRIS, Charles.
Son of the above. Neat workmanship. Well-cut scroll. Sound-holes
not well formed. Yellow varnish. Worked for a short time for John
Hart.
HART, John Thomas,
born December 17, 1805, died January 1, 1874. He was articled to
Samuel Gilkes in May, 1820, of whom he learned the mechanical branch
of his profession. He afterwards centred his attention upon the
peculiar characteristics of the Cremonese and Italian Violin-makers
generally, and in a comparatively brief space of time obtained an
extensive acquaintance in that direction. His unerring eye and powerful
memory of instruments once brought under his notice secured for
him the highest position among the connoisseurs of his time. Commencing
business at a period when the desire to possess instruments by the
famous Italian makers was becoming general among amateurs, and being
peculiarly fortunate in securing an early reputation as a judge
of them, he became the channel through which the greater part of
the rare Italian works passed into England, and it has frequently
been said that there are very few distinguished instruments in Europe
with which he was unacquainted. Among the remarkable collections
that he brought together may be mentioned that of the late Mr. James
Goding, the remnant of which was dispersed by Messrs. Christie and
Manson in 1857; the small but exquisite collection of Mr. Charles
Plowden, consisting of four Violins of Stradivari and four of Guarneri,
with other instruments of less merit, the whole of which again passed
into Mr. Hart's possession upon the death of their owner; and, lastly,
a large portion of the well-known collection of the late Mr. Joseph
Gillot, sold by Christie and Manson shortly after the famous sale
of pictures belonging to the same collector. HAYNES, Jacob, London,
1746. Copied Stainer. The style resembles that of Barrett. HEESOM, Edward, London,
1748. Copied Stainer. HILL, Joseph, London,
1715-84. Pupil of Peter Wamsley. His Violoncellos and Tenors are
well-made instruments. HILL, William, London,
1741. Son of the above. Very good work. HILL, Joseph, London,
1800-40. Son of the above. HILL, Lockey, London,
1800-35. Brother of the above. Made many excellent instruments. HILL, William Ebsworth,
London, 1817-95. Son of Lockey Hill. Made several instruments in
his younger days, but, like the rest of our English makers, he long
since discovered that new work was unremunerative, and turned his
attention to repairing and dealing in old instruments, and became
the founder of the well-known firm of W. E. Hill and Sons, of Bond
Street. He exhibited at the Exhibition of 1862 a Violin and Tenor,
thus showing that Violin-making was not quite extinguished in England. HOLLOWAY, J., London,
1794. HUME, Richard, Edinburgh,
16th century. A maker of Lutes, &c. JAY, Henry, London,
17th century. Maker of Viols, which are capital specimens of the
work of the period. The varnish is excellent. JAY, Thomas, London.
Related to the above. Excellent work. JAY, Henry, London,
about 1744-77. A maker of Kits chiefly. At this period these juvenile
Violins were in much demand by dancing-masters. A few years ago
a very choice collection of these instruments was made by an Irish
gentleman residing at Paris, who obtained specimens from all parts
of Europe. Henry Jay also made Violoncellos, some of which have
the names of Longman and Broderip on the back. JOHNSON, John, London,
1750. The Violins bearing his label are dated from Cheapside. Johnson
was a music and musical instrument seller. In "The Professional
Life of Dibdin," written by himself, we have the following
reference to this City music-seller: "My brother introduced
me to old Johnson, who at that time kept a capital music-shop in
Cheapside.1 I soon, however, grew tired of an attendance
on him. He set me down to tune Harpsichords, a mere mechanical employment,
not at all to my taste." "I saw plainly that I might have
screwed up Harpsichords in old Johnson's shop to all eternity, without
advancing my fortune; and as to the songs and sonatas that I brought
him for sale, they had not been performed at the theatres nor Vauxhall,
nor any other place, and Johnson would not print them." "The
Thompsons, however, of St. Paul's Churchyard, published six ballads
for me, which sold at three-halfpence a-piece, and for the copyright
of which they generously gave me three guineas." Though we
may not feel disposed to apply the term "generous" to
a payment of half-a-guinea for a Dibdin ballad, yet in all probability
we are indebted to the Thompsons for this particular recognition
of merit. Happily true genius, when in straits, generally finds
relief. Were it otherwise, and had the Thompsons been as deaf to
Dibdin as John Johnson appears to have been, "Tom Bowling,"
"Poor Jack," and many other compositions of sterling merit,
might never have been written.2 1 Dibdin's brother
was captain of a merchant vessel, and was intimate with Johnson
the music-seller. On the death of Captain Dibdin his brother composed
"Tom Bowling," the music and words of which bespeak the
fraternal love of the composer. 2 Dibdin was evidently
discouraged in consequence of Johnson's refusal to publish his songs:
he says, "After I had broken off with Johnson, I had some idea
of turning my thoughts to merchants' accounts—the very last
thing upon earth for which I was calculated." KENNEDY, Alexander,
London, 1700-86. Was a native of Scotland. He was the first maker
of Violins in his family, which was connected with the manufacture
for nearly two centuries. KENNEDY, John, London,
born 1730; died 1816. Nephew of Alexander Kennedy. Made Violins
and Tenors. KENNEDY, Thomas,
London, born 1784; died about 1870. Son of the above. Probably made
more instruments than any English maker, with the exception of Crask. LENTZ, Johann Nicolaus,
London, 1803. He used mostly one kind of wood, viz., close-grained
maple. Varnish nearly opaque. LEWIS, Edward, London,
1700. The work is well executed throughout, and the varnish superior. LISTER, George, 18th
century. LONGMAN AND BRODERIP,
Cheapside, London, about 1760. They were music-publishers and instrument-sellers,
and were not Violin-makers. Benjamin Banks, Jay, and others, made
many of the instruments upon which the name of Longman is stamped.
Muzio Clementi was at one time a partner in the firm. The business
ultimately passed to Collard and Collard. LOTT, John Frederick,
1775-1853. Was a German by birth. He was engaged in the cabinet
business early in life. He was induced by Fendt to turn his attention
to making Violins, and ultimately obtained employment under Thomas
Dodd, making many of the Violoncellos and Double-Basses that carry
the label of Dodd within them. His work was of a most finished description.
His Double-Basses are splendid instruments, and will bear comparison
with Italian work. His varnish was far from equal to his finish.
The time he spent in making these instruments was double that which
any other English maker expended over similar work. There is not
a single portion of any of his Double-Basses that has been carelessly
made; the interior is as beautifully finished as the exterior. The
machines on many of his Basses were made by himself—a very
unusual circumstance. The scrolls are finely cut. He was certainly
the king of the English Double-Bass makers. LOTT, George Frederick,
London, born 1800; died 1868. Son of the above. Many years with
Davis, of Coventry Street. Was an excellent judge of Italian instruments,
and a clever imitator. LOTT, John Frederick,
London, younger brother of the above, died about 1871. Was articled
to Davis. Has made many clever imitations. He was also an ardent
lover of Cremonese instruments, and thoroughly understood their
characteristics. His career was both chequered and curious, sufficiently
so, indeed, to cause our eminent novelist, Charles Reade, to make
it the subject of "Jack of all Trades: a Matter-of-Fact Romance."
Jack Lott (as he was familiarly styled) therefore shares with Jacob
Stainer the honour of having supplied subject-matter for writers
of fiction. It must, however, be said that whilst Dr. Schuler's
"Jacob Stainer" is mainly pure fiction, "Jack of
all Trades" is rightly entitled "a matter-of-fact romance."
I have many times heard John Lott relate the chief incidents so
graphically described by Charles Reade. MACINTOSH, Dublin.
Succeeded Perry and Wilkinson. Died about 1840. MARSHALL, John, London,
1750. MARTIN, ——,
London, about 1790. MAYSON, Walter H.,
Manchester, 1835-1904. A prolific maker. His later work is highly
spoken of. MEARES, Richard,
about 1677. Maker of Viols. MIER, ——,
London, about 1786. MORRISON, John, London,
about 1780-1803. NAYLOR, Isaac, Headingly,
near Leeds, about 1778-92. NICHOLS, Edward,
18th century. NORBORN, John, London,
about 1723. NORMAN, Barak, London,
1688-1740. The instruments of this maker are among the best of the
Old English school. His instructor in the art of Viol and Violin-making
is unknown, but judging from the character of his work it is very
probable he learned from Thomas Urquhart. This opinion is strengthened
upon examining his earliest instruments. We there find the same
peculiarities which mark the individuality of Urquhart. Later in
life he leaned much to the model of Maggini. During his early
years he was much esteemed as a maker of Viols, many of which have
all the marks of careful work upon them. On all of these instruments
will be found his name, surrounded with a design in purfling, under
the finger-board, or his monogram executed in purfling. The same
trade token will be found in his Violoncellos. All endeavours to
discover any existing English Violoncello, or record of one, anterior
to Barak Norman, have failed, and, consequently, it may be assumed
that he was the first maker of that instrument in England. Here,
again, is evidence of his partiality for the form of Maggini, as
he copied this maker in nearly all his Violoncellos. All the Violoncellos
of Barak Norman have bellies of good quality; the modelling is executed
skilfully, due care having been observed in leaving sufficient wood.
His Tenors are fine instruments. Many of these were made years before
he began the Violoncellos—a fact which satisfactorily accounts
for the marked difference in form peculiar to them. The build is
higher, and the sound-hole German in character; the varnish is very
dark. About the year 1715 Barak Norman entered into partnership
with Nathaniel Cross, carrying on the joint business at the sign
of the Bass Viol, St. Paul's Churchyard. In a Viol da Gamba which
belonged to Walter Brooksbank, Esq., of Windermere, is a label in
the handwriting of Nathaniel Cross, in which he adds the power of
speech to the qualities of the quaint Gamba; the words are, "Nathaniel
Cross wrought my back and belly," the sides and scroll being
the work of his partner. NORRIS, John, London,
born 1739; died 1818. Articled to Thomas Smith, the successor of
Peter Wamsley. Similar work to that of Thomas Smith. He became a
partner of Robert Barnes. PAMPHILON, Edward,
London, 17th century. The Violins of this maker were formerly much
prized. The model is very high, and the appearance somewhat grotesque.
It is to be regretted that the splendid varnish often found on these
instruments was not put upon better work. PANORMO, Joseph,
London. Son of Vincent Panormo. His work was excellent. His Violoncellos
are decidedly superior to his Violins. PANORMO, George Lewis,
London. Brother of the above. Made Violins of the Stradivari pattern. PANORMO, Louis, London.
Made Guitars chiefly. PARKER, Daniel, London,
18th century. This is another maker of the English school, who was
possessed of exceptional talent, and whose instruments are well
worthy of attention from those in search of good Violins at a moderate
cost. To Parker belongs, in conjunction with Benjamin Banks, the
merit of breaking through the prejudice so long in favour of preference
for the Stainer model. The dates of his
instruments extend from the year 1740 to 1785. He left his Violins
thick in wood, which has certainly enhanced their value now that
time has ripened them. He used excellent material, which is often
very handsome. The varnish is of a mellow quality, and fairly transparent.
A large number of these Violins have been passing under other makers'
names, and have been but little noticed. PEARCE, James, London,
18th century. PEARCE, W., London,
contemporary. PEMBERTON, Edward,
London, 1660. This maker has been often mentioned as the author
of a Violin said to have been presented to the Earl of Leicester
by Queen Elizabeth, and to suit this legend Pemberton's era has
been put back a century. The date given above will be found in the
Violins of this maker. PERRY AND WILKINSON,
Dublin, 17— to 1830. The instruments bearing the labels of
these makers are frequently excellent in tone, material, and finish. POWELL, Thomas, London,
18th century. PRESTON, London,
about 1724. Appears to have used his trade label in the instruments
he sold, made by makers he employed. PRESTON, John, York,
18th century. RAWLINS, Henry, London,
about 1781. He appears to have been patronised by Giardini, the
Violinist, according to the label here given. Giardini held the
post of leader at the Italian Opera at this period. RAYMAN, Jacob, London,
17th century. The subject of this notice was probably a German,
from the Tyrol, who settled in England about 1620, and may be considered
as the founder of Violin-making in this country, there being no
trace of any British Violin-maker previous to that time. His work
is quite different from that of the old English Viol-makers. The
instruments of Rayman are of a somewhat rough exterior, but full
of character. The form is flat, considering the general style of
the work. The sound-holes are striking, although not graceful in
any way. The scroll is diminutive, but well cut. The varnish is
very fine. In the catalogue of the effects of Tom Britton, mention
is made of "an extraordinary Rayman." RICHARDS, Edwin,
London, contemporary. Maker and repairer. ROOK, Joseph, Carlisle,
about 1800. ROSSE (or Ross),
John, Bridewell, London, about 1562. Made Viols and Bandoras. ROSS, John, London,
about 1596. Son of the above. Maker of Viols. The varnish is excellent
in quality. SHAW, London, 1655.
Viol maker. SIMPSON, London,
1785. SMITH, Henry, London,
1629. Maker of Viols. SMITH, Thomas, London.
Pupil of Peter Wamsley, and his successor at the Harp and Hautboy. SMITH, William, London,
about 1770. TARR, William, Manchester.
Made many Double-Basses from about 1829. TAYLOR, London, about
1800. A maker of much merit. Instruments of the character of Panormo. THOMPSON, London,
1749. THOROWGOOD, Henry,
London. Little known. TILLEY, Thomas, London,
about 1774. TOBIN, Richard, London,
1800. Pupil of Perry, of Dublin. His instruments are much appreciated
by the best judges. In cutting a scroll he was unequalled amongst
English makers. TOBIN, London. Son
of the above. URQUHART, London,
17th century. Nothing is known concerning the history of this excellent
maker. The style may be considered as resembling that of Jacob Rayman,
and it is possible he worked with him. His varnish is equal to that
on many Italian instruments. VALENTINE, William,
London, died about 1877. Made many Double-Basses for Mr. Hart, which
are highly valued. WAMSLEY, Peter, London,
18th century. One of the best English makers. His copies of Stainer
are very superior. WISE, Christopher,
London, about 1650. Yellow varnish, neat workmanship, flat model,
small pattern. WITHERS, Edward,
Coventry Street. Succeeded William Davis. WITHERS, Edward.
Son of the above. Wardour Street, Soho. Was instructed by John Lott. YOUNG, London, about
1728. Lived in St. Paul's Churchyard. Purcell has immortalised father
and son in the first volume of his Catches.
CONTENTS
SECTION
V.—THE GERMAN SCHOOL AND MAKERS SECTION
VI.—THE ENGLISH SCHOOL AND MAKERS SECTION
VII.—THE VIOLIN AND ITS VOTARIES. SKETCH OF THE PROGRESS
OF THE VIOLIN
|