Vallotti temperament

Círculo de quintas correspondiente al temperamento de Vallotti de 1/6 de coma, según Tartini.

The circulating temperament today referred to as Vallotti temperament (or simply Vallotti, Vallotti-Barca, Vallotti-Tartini, or Vallotti-Young) is a shifted version of Young's second temperament. Its attribution to the 18th-century organist, composer, and music theorist, Francesco Vallotti is a mistake, since there is no evidence that he ever suggested it. It is however audibly indistinguishable from a slightly different temperament that was in fact devised by Vallotti.

Vallotti's description of his temperament appears in book 2 of his treatise, Della scienza teorica e pratica della moderna musica (On the theoretical and practical science of modern music). Although he stated that he had developed his theoretical system—presumably including the details of his temperament—by 1728, the first book of his treatise was not published until 1779, the year before he died. At the time of his death, the other three books had not been published, and remained only in manuscript form until an edition of all four books was published in 1950, under the title Trattato della moderna musica (Treatise on modern music).^[1]

Vallotti's temperament received very little attention during his lifetime and for some time thereafter.^[2] In a treatise published in 1754,^[3] Vallotti's friend and colleague Giuseppe Tartini praised the former's approach to temperament, and outlined some of its features, but without giving sufficient detail for the temperament itself to be identified.^[4] In 1781, the mathematician William Jones noted Tartini's preference for Vallotti's temperament and gave a similarly vague and unspecific description.^[5]

The temperament originally devised by Vallotti has six fifths tempered by 1⁄6 of a syntonic comma, five perfectly just, and one tempered by a schisma. In a manuscript which remained unpublished until 1987,^[6] the Italian chemist and musical theorist, Alessandro Barca, proposed that this latter fifth be widened by 5⁄6 of a schisma, and all the pure fifths be narrowed by 1⁄6 of a schisma. Barca's version thus has six fifths tempered by 1⁄6 of a syntonic comma, and six tempered by 1⁄6 of a schisma. In the temperament now commonly misattributed to Vallotti, the odd fifth out in his original is widened by a full schisma, and each of the six tempered fifths is narrowed by a further 1⁄6 of a schisma. This modern version thus has six fifths tempered by 1⁄6 of a Pythagorean comma, and six perfectly just. More recently, Owen Jorgensen has proposed a version of Vallotti's temperament in which the beating frequencies of the tempered fifths, rather than their sizes, are chosen to be equal. In practice, none of these four versions is audibly distinguishable from any of the others,^[7] because no interval in any of them differs from the corresponding interval in any of the other three by more than two cents.

In the circulating temperament today commonly misattributed to Vallotti, each of the fifths B-F♯, F♯-C♯, C♯-G♯, G♯-E♭, E♭-B♭, and B♭-F are perfectly just, while the fifths F-C, C-G, G-D, D-A, A-E, and E-B are each 1⁄6 of a Pythagorean (ditonic) comma narrower than just.^[8] The exact and approximate numerical sizes of these fifths, in cents, are given by:

f₁	Def=	1200 ( log₂(3) − 1) ≈ 701.96	(perfectly just)
f₂	Def=	2600 − 1200 log₂(3) ≈ 698.04	(narrower than just by 1⁄6 of a ditonic comma)

If s_j Def= f_j − 600 for j = 1,2, the sizes of the major thirds in this temperament are:^[9]

Major third	F-A, C-E, G-B	D-F♯, B♭-D	A-C♯, E♭-G	E-G♯, G♯-C	B-E♭, F♯-B♭, C♯-F
Width exact approx.	4 s₂ 392.18	3 s₂ + s₁ 396.09	2 s₂ + 2 s₁ 400 (exactly)	s₂ + 3 s₁ 403.91	4 s₁ 407.82
Deviation from just	+5.9	+9.8	+13.7	+17.6	+21.5

The following table gives the pitch differences in cents between the notes of a chromatic scale tuned with this temperament and those of one tuned with equal temperament, when the note A of each scale is given the same pitch.^[10]

Note	E♭	B♭	F	C	G	D	A	E	B	F♯	C♯	G♯
Difference from equal temperament	+3.9	+5.9	+7.8	+5.9	+3.9	+2.0	0	-2.0	-3.9	-2.0	0	+2.0

This temperament is merely a shifted version of Young's second temperament, which also has six consecutive pure fifths and six tempered by 1⁄6 of a Pythagorean comma. In Young's second temperament, however, the sequence of tempered fifths starts from the note C, rather than from F, as they do in the temperament today commonly misattributed to Vallotti.^[11]

Other versions

Vallotti's original

In the original description of his temperament, Vallotti made each of the fifths B-F♯, F♯-C♯, C♯-G♯, G♯-E♭, and E♭-B♭ perfectly just, just as in the modern version, but rather than making the fifths F-C, C-G, G-D, D-A, A-E, and E-B narrower than just by a 1⁄6 of a Pythagorean comma, he had narrowed them by only 1⁄6 of a syntonic comma. This left the remaining fifth, B♭-F, narrower than just by a schisma.^[12] The exact and approximate numerical sizes of these latter fifths, in cents, are given by:

f₃	Def=	200 ( 2 log₂(3) + log₂(5) – 2 ) ≈ 698.37	(narrower than just by 1⁄6 of a syntonic comma)
f₄	Def=	1200 ( 14 – 7 log₂(3) – log₂(5) ) ≈ 700.00	(narrower than just by a schisma)

If s₁ is defined as above, and s_j Def= f_j − 600 for j = 3,4, the sizes of the major thirds in this temperament are:

Major third	F-A, C-E, G-B	D-F♯	A-C♯	E-G♯	B-E♭, F♯-B♭	C♯-F	G♯-C	E♭-G	B♭-D
Width exact approx.	4 s₃ 393.48	3 s₃ + s₁ 397.07	2 s₃ + 2 s₁ 400.65	s₃ + 3 s₁ 404.24	4 s₁ 407.82	s₄ + 3 s₁ 405.87	s₃ + s₄ + 2 s₁ 402.28	2 s₃ + s₄ + s₁ 398.70	3 s₃ + s₄ 395.11
Deviation from just	+7.2	+10.8	+14.3	+17.9	+21.5	+19.6	+16.0	+12.4	+8.8

The following table gives the pitch differences in cents between the notes of a chromatic scale tuned with this temperament and those of one tuned with equal temperament, when the note A of each scale is given the same pitch.

Note	E♭	B♭	F	C	G	D	A	E	B	F♯	C♯	G♯
Difference from equal temperament	+4.6	+6.5	+6.5	+4.9	+3.3	+1.6	0	-1.6	-3.3	-1.3	+0.65	+2.6

Barca's suggested modification

In an 18th-century work, which remained unpublished until 1987, Alessandro Barca suggested that the schisma discrepancy which Vallotti had left to fall entirely in the single fifth, B♭-F, be instead spread amongst the six fifths B-F♯, F♯-C♯, C♯-G♯, G♯-E♭, E♭-B♭, and B♭-F, thus making them each narrower than just by the negligible quantity 1⁄6 of a schisma (about 1⁄3 of a cent). The exact and approximate numerical size of these fifths, in cents, is given by:

f₅

200 ( 9 – 2 log₂(3) – log₂(5) ) ≈ 701.63

(narrower than just by 1⁄6 of a schisma)

If s₃ is defined as above, and s₅ Def= f₅ − 600, the sizes of the major thirds in this temperament are:

Major third	F-A, C-E, G-B	D-F♯, B♭-D	A-C♯, E♭-G	E-G♯, G♯-C	B-E♭, F♯-B♭, C♯-F
Width exact approx.	4 s₃ 393.48	3 s₃ + s₅ 396.74	2 s₃ + 2 s₅ 400 (exactly)	s₃ + 3 s₅ 403.26	4 s₅ 406.52
Deviation from just	+7.2	+9.5	+13.7	+17.0	+20.2

The following table gives the pitch differences in cents between the notes of a chromatic scale tuned with this temperament and those of one tuned with equal temperament, when the note A of each scale is given the same pitch.

Note	E♭	B♭	F	C	G	D	A	E	B	F♯	C♯	G♯
Difference from equal temperament	+3.3	+4.9	+6.5	+4.9	+3.3	+1.6	0	-1.6	-3.3	-1.6	0	+1.6

Jorgensen's version with equal-beating fifths

One of the leading experts on keyboard construction and tuning, Owen Jorgensen, contended that tempering fifths by precisely the same amount on keyboards—with the possible exception of the organ—was beyond the capabilities of tuning practices used before the twentieth century,^[13] and that the vast majority of keyboard tuners, when tuning by ear before the development of twentieth century tuning techniques, would have judged two adjacent or overlapping fifths to be the same whenever they beat at the same rate.^[14]

Jorgensen gave two sets of instructions for tuning Valotti's temperament in a way which he considered representative of what he believed would have been the results achieved by 18th- and 19th-century tuners. The first used a bearing plan for the octave F₃ to F₄,^[15] the second, a bearing plan for the higher octave, F₄ to F₅.^[16] In the first, middle C (C₄) is tuned to a standard pitch of $220 4 \sqrt 2$ Hz, all octaves, and the fifths B–F♯, F♯–C♯, C♯–G♯, E♭–B♭ and B♭–F are tuned just, while the fifths F₃–C₄, C₃–G₃, G₃–D₄, D₃–A₃, A₃–E₄, and E₃–B₃ are tuned narrow, all with a beat rate of 1.1 Hz. The amounts by which these tempered fifths are narrow range from 2.9 cents for A–E to 4.9 cents for C–G, and average to 3.8 cents, slightly less than a sixth of the Pythagorean comma. As a consequence, the diminished sixth G♯–E♭, which is required to be a perfectly just fifth in Vallotti proper, turns out to be tempered narrow by 0.6 cents in this version of Jorgensen's. The sizes of its major thirds in cents are:^[17]

Major third	F-A	C-E	G-B	D-F♯	A-C♯	E-G♯	B-E♭, F♯-B♭, C♯-F	G♯-C	E♭-G	B♭-D
Width (approx.)	391.74	392.48	393.46	396.71	401.05	403.95	407.21	403.58	399.32	396.08
Deviation from just	+5.4	+6.2	+7.1	+10.4	+14.7	+17.6	+20.9	+17.3	+13.0	+9.8

The following table gives the pitch differences in cents between the notes of a chromatic scale tuned with Jorgensen's equal-beating version of Vallotti temperament and those of one tuned with equal temperament, when the note C₄ of each scale is given the same pitch, $220 4 \sqrt 2$ Hz.^[18]

Note	E♭	B♭	F	C	G	D	A	E	B	F♯	C♯	G♯
Difference from equal temperament	-2.2	-0.3	+1.7	0	-2.9	-4.2	-6.6	-7.5	-9.4	-7.5	-5.5	-3.5

Notes

↑ Damschroder and Williams (1990, p.365); Hansell (2001, 2007).
↑ Barbieri (1982, pp.63, 65). Barbieri quotes the chemist and musical theorist, Alessandro Barca, writing sometime after Vallotti's death, as saying that his temperament had been little more than merely referred to in his unpublished writings.
↑ Trattato di musica secondo la vera scienza dell' armonia (Tartini, 1754)
↑ Tartini (1754, p.100). Benjamin Stillingfleet translates the relevant passage from Tartini's treatise as follows (Stillingfleet, 1771, p.35):
"and I infinitely applaud the opinion of P. Vallotti, our organ-master, as the most reasonable of all. He says, that you ought to give to the white keys of the organ all their natural perfection –, both because they are the natural notes of the diatonic genus, and because in church-music the greatest use is made of them; throwing thus the greatest imperfection upon those black keys, which are most remote from the diatonic scale, and which are hardly ever used."
The original Italian reads (Tartini, 1754, p.100):
"ed io lodo infinitamente il sentimento del Padre Valloti nostro Maestro come il più ragionevole di tutti, perchè il più prudente. Egli, dice, che si deve lasciare a' tasti bianchi dell' organo tutta la loro naturale perfezione; sì perchè sono li naturali del Genere diatonico; sì perchè di quelli nel servigio Ecclesiastico sé ne fa il maggior uso: riducendo la massima imperfezione a que' tasti neri, che fono i più lontani dal Genere diatonico, e di quasi niun' uso."
↑ Jones (1781, pp.325–326).
↑ When it was published in Patrizio Barbieri's Acustica accordatura e temperamento nell’Illuminismo veneto. Con scritti inediti di Alessandro Barca, Giordano Riccati e altri autori (Barbieri, 1987).
↑ Except by carefully timing the beats generated by some non-just intervals.
↑ Donahue (2005, p.28)
↑ Jorgensen (1991, Table 51-1, p.180).
↑ Jorgensen (1991, Table 52-1, p.185).
↑ Donahue (2005, pp.28–9 )
↑ Barbieri (1982, p.63); Di Veroli (2013, p.125)
↑ Jorgensen (1991, pp.44–45, 175, 310–11). Although the only specific temperaments Jorgensen mentions in this connection are 12-tone equal and quarter-comma meantone, his comment on (on p.45) that "[t]wentieth-century acoustic knowledge and test-interval techniques are required for tuning any regular temperament by ear" implies a belief that the same limitation applied to the tempering of fifths precisely by any specified amount.
↑ Jorgensen (1991, pp.44, 310). Sturm (2011) and Di Veroli (2013, pp.146-48) have dismissed these contentions of Jorgensen's as unfounded speculation.
↑ Jorgensen (1991, pp.68–73). Notes are specified here using scientific pitch notation. F₃ is the highest F below middle C, and F₄ is the next above it.
↑ Jorgensen (1991, pp.176–77).
↑ Jorgensen (1991, Table 20-1, p.65).
↑ Jorgensen (1991, Table 22-1, p.72). Jorgensen's table 22-1 is given for an equal temperament tuned to a pitch 6.58097 cents lower than standard, so that the pitch of its A will coincide with that of his equal-beating Vallotti. The table given here is for an equal temperament at standard pitch. Its entries are those of Jorgensen's Table 22-1 decreased by 6.58097 cents and rounded to one decimal place.