The 'Bach Organ' at Arnstadt
Home Up Other Publications Contact Me About the Author About this Website Prog Organ - a VPO Sitemap What they say

 

  The 'Bach Organ' at Arnstadt  

 

Colin Pykett

 

"the most productive and helpful coming together of historical study and live playing happens when what one is studying is not books or even scores but an historic instrument"
Peter Williams [1]

 

Posted: 26 July 2021
Revised: 27 July 2021
Copyright C E Pykett

 

Abstract. The young J S Bach was appointed organist at what is now known as the Bachkirche in Arnstadt in 1703, leading to today's familiarity coexisting with persistent misunderstandings about the instrument. For more than three centuries writers have revealed uncertainties over such basic matters as its stop list, leading some to propose dispositions which could not even have fitted on the jambs of the still-extant console. Thus more subtle matters such as how music might have been registered on it have not been well addressed. At one extreme it can be argued that the tonal design of the instrument defies logic compared with contemporary Werkprinzip organs to the north, and to counter this it is necessary to examine the rapid changes in musical taste and registrational styles taking place post-1700.

This article examines these and other topics against the backdrop of a great man just setting out on his career. It draws on the meticulous research undertaken when the organ was reconstructed in 1999, which seems not to have been well publicised outside Germany. 

 

 

Wender's 'Bach Organ' at Arnstadt (1703) as recreated by Hoffmann (1999)
(Creative Commons Universal Public Domain Dedication)

 

Contents
(click on the headings below to access the desired section)

 

Introduction

The Wender organ

The Hesse organ

The Steinmeyer organ

The Hoffmann organ

Concluding remarks

Notes and references

 

 

Introduction

 

Most organists and others with an interest in J S Bach are aware of the so-called 'Bach Organ' at Arnstadt in Thuringia, and even tourists leave the area with some recognition of its significance. Yet this familiarity is partly a façade concealing an historical patchwork full of holes held together at least as much by anecdote, myth and legend as by reality. Too much of the associated literature is vague and peppered with misunderstandings and errors, not a little of which has been amplified by the inconsequential internet ephemera of recent times. When pursuing more serious research further difficulties are faced by those who do not read German or who lack ready access to information not in the public domain. Against this background, the present article therefore tries to rescue the organ and the church (their stories are inextricably linked) from a rather untidy situation. In doing this it is fortunate that we have a detailed account of the reconstruction of the building and its organ which took place between 1997 and 2000 [3], incorporating a history of both going back to their earliest days. Although recommended reading, it is in German and as far as I know it has not been published widely, thus providing additional justification for this article.

 

 

The Wender organ

 

So let us begin with J S Bach's appointment as organist at Arnstadt in his late teens in 1703, the year that J F Wender's entirely new organ had been inspected and approved by the youngster himself. The church was then called the Neuekirche because its predecessor, Bonifaciuskirche or St Boniface, had been destroyed by fire in 1581. Today it is called the Bachkirche for obvious reasons, having taken this name in 1935. In Bach's time the church was as lofty as it is now, thus it would have enjoyed much the same impressive acoustic. In such a large building it might therefore seem odd that Wender's organ was so small and banished to a position high on the top gallery, crammed into a barely adequate space under the vaulting such that insufficient height was available for some stops (see the picture above). The longest pipes of the 16 foot pedal Posaunen Bass and 8 foot Trompet on the Oberwerk both had fractional length resonators for this reason [3]. Furthermore, because a manual 16 foot flue was similarly out of the question, a curious sub-quint principal at 5 1/3 foot pitch was presumably included on the Oberwerk so that a fake suboctave effect might be obtained in conjunction with the unison stops. But this begs the question as to why Wender did not simply provide a stopped 8 foot rank to generate real 16 foot tone instead if that is what he wanted?

 

 

Figure 1. The actual console of the original Wender organ of 1703

 

But we are getting ahead of ourselves in discussing the sounds of the instrument before we have even had a peek at the stop list. Establishing the original disposition of any organ is often difficult and sometimes impossible, but in this case the original contract drawn up with Wender in 1699 still exists. However some modifications were apparently made during its construction or shortly afterwards (e.g. the Nachthorn on the Brustwerk was added later). Nevertheless, the precious original console with its drawstops and some labels also still exists in the town museum (Figure 1) and consequently Hoffmann was able to reconstruct the instrument in 1999 with some confidence to the disposition shown in Table 1. 

 

Pedal 

 

 

Brustwerk

 

 

Oberwerk

 

Sub Baß

16

 

Still gedackt

  8

 

Principal

8

Principal Baß

  8

 

Principal

  4

 

Viol di Gamba

8

Posaunen Baß

16

 

Nachthorn

  4

 

Gemshorn

8

Cornet Baß

  2

 

Spitz flöte

  4

 

Quinta dena

8

 

 

 

Quinte

  3 

 

Grob gedackt

8

Ow - Pedal

 

 

Sesquialtera doppelt

  II

 

Quinta

6

 

 

Mixtur 3 fach

 

 

Octava 

4 

 

 

 

 

 

Mixtur 4 fach

 

 

 

 

 

 

Cymbel doppelt

II 

Trompet 8

Tremulant

Cymbalstern (C) Bw - Ow
Cymbalstern (G)

 

 

Table 1. Neuekirche, Arnstadt (Wender, 1703; reconstructed by Hoffmann, 1999)

 

It is worth pointing out that some other sources contain errors. Williams and Owen [4] gave both manuals a 2 foot stop, they gave the pedal a non-existent Violon but omitted the Cornet, they were unclear about the couplers, they wondered whether an 8 foot Höhl Flute appeared on the Brustwerk and whether the Quintadena was at 16 foot pitch. Sumner [5] and Schweitzer [6], probably following Bach's 19th century biographer Spitta, both put a 4 foot flute on the pedal. Wills [7] refers to 'the Pedal of five stops' (sic) without listing them, perhaps following the flawed assumptions of Schweitzer and Sumner. Some authors have even included more stops than could have been accommodated on the jambs, a lapse that cannot be overlooked easily because the oft-pictured original console still exists and has therefore always been available for study, the Cold War years notwithstanding (the photograph in Figure 1 appeared widely in Western publications in the 1960s, for example).

Note the presence of five unison stops on the Oberwerk besides the Trompet, a Quinta of 6 (i.e. 5 1/3) foot pitch, a single 4 foot stop and then nothing between it and the mixtures. The smaller Brustwerk is conceived similarly, though related to a 4 foot Principal. The difference in pitch datum is also reflected in the mixture compositions on the two manuals - that on the Oberwerk starts at 2 foot pitch and on the Brustwerk at 1 foot. The Cymbel is also a 1 foot mixture. All these are quint mixtures containing no third-sounding ranks. However the Sesquialtera does, of course, have a tierce rank at the interval of a 17th above unison together with a 12th. There is therefore a generous helping of twelfth tone on the Brustwerk in that the Quinte by itself was probably fairly prominent [3], as it is today in the reconstructed instrument, and in combination with the Sesquialtera there would have been further reinforcement of this pitch. Again in terms of pitch, the pedal organ offers nothing between the 8 foot Principal and the 2 foot Cornet, a trumpet-toned reed whose diminutive pipes stood on the soundboard one-on-one in front of their huge cousins in the Posaune rank. The tonal palette is augmented by not one but two fairly loud Cymbalsterne employing gongs struck by hammers, activated by separate drawstops and tuned to broken chords in C and G major respectively.

The disposition is therefore curious in some ways. Although replete with many different unison tone colours for so small an instrument, we have noted that the vertical choruses on all divisions are incomplete and particularly so for the pedal organ. Even something so basic as a 'gap' registration at 8 foot and 2 foot on the manuals is not catered for here. Being particularly unkind, one might regard the instrument almost as a baroque 'octopod' apart from the redemption of its mixtures. It was very different to organs of similar size and age found further north, of which a well known example is Arp Schnitger's instrument of 1680 at St John's monastery in Hamburg. Now at Cappel, this famous instrument still exists and its stop list is in Table 2.

 

Hauptwerk

 

 

Rückpositiv

 

 

Pedal

 

Quintad

16

 

Quintad

  8

 

Untersatz

 16

Principal

  8

 

Gedact

  8

 

Octava

   8

Hollflöit

  8

 

Principal

  4

 

Octava

   4

Octava

  4

 

Flöit

  4

 

Nachthorn

   2

Spitzflöit

  4

 

Octava

  2

 

Rauschpfeife

  II

Nasat

  3

 

Sifflöit

1 1/2 

 

Mixtur

IV-VI

Gemshorn   2 Sesquialter  II Posaune  16
Rauschpfeife  II Tertian  II Trompet   8
Mixtur V-VI Scharff IV-VI Cornet   2

Zimbel

III 

 

Dulcian

16

 

 

 

Trompet 

  8

 

 

 

 

 

 

 

Tremulant

 

 

 

RP - HW Cymbalstern

 

Table 2. Johannisklosters, Hamburg, now at Cappel (Arp Schnitger, 1680) 

 

The differences between the two organs could scarcely be more striking. Schnitger's had completely developed chorus work on all three divisions whereas Wender's did not; Schnitger's limited range of unison tonalities was subservient to this whereas Wender explored diversity of timbre to a far greater degree; possibilities for synthetic tone building using mutation stops were almost negligible on Wender's organ compared with the riches on Schnitger's; Schnitger included real 16 foot tone on both manual divisions while Wender provided none; unlike Wender, Schnitger provided no pedal coupler because he did not need to with so complete a pedal organ; and Schnitger included a 'toy' stop but Wender went further and installed two.

Conventional wisdom from some quarters suggests that improvements in organ building technology partly explain these differences. For instance, Schnitger's organs show clearly the early design precepts of Michael Praetorius (1571-1621) who laid down rules in his Syntagma Musicum such as that which deprecated drawing more than one stop at each pitch. This was partly to prevent an unacceptable load being thrown onto the unfortunate organ blowers and partly to maintain the desired wind pressure and thus preserve good enough mutual tuning between ranks. In his writings around the time that the Arnstadt organ was being built Andreas Werckmeister (1645-1706) mandated the same thing for the same reasons. Yet we are sometimes expected to believe that, magically, Wender was able to do things differently at Arnstadt. Some say that his organ positively encouraged players to draw many unison stops together, as Bach was said to have done (if not here, then elsewhere), thereby multiplying even further the already generous provision of many tone colours. However it is unclear to me why it is thought that the rudimentary winding system of Wender's little organ in a small and remote village could somehow have coped with this, whereas that of the great Schnitger only twenty years earlier in the sophisticated city of Hamburg could not. So I am afraid this hypothesis, like others of similar ilk, seems to wither on the vine. It is easy to draw facile conclusions about the winding of old organs when you have just pressed the starter button on a modern console, but it is unscholarly and definitely anachronistic. Wind provision for organs in the pre-industrial era is not adequately explored nor understood beyond a few obvious qualitative observations, and to justify this assertion I only need to point to a study of the physics of human-powered organ blowing which I undertook recently [8]. If a satisfactory literature had existed on the subject it would have been unnecessary for me to go to these lengths. Now that we have Hoffmann's reconstructed instrument provided with both electric and hand blowing, perhaps a few simple experiments would yield interesting results. The obvious one would be to see whether human blowers, often just children in Bach's day, could provide enough wind for the many unison stops drawn together. If so, for how long could they maintain the effort, and would the mutual tuning of the ranks suffer?

The more likely reality is that, in these two instruments, we see a reflection of rapidly changing musical tastes in the latter years of the seventeenth century and during the first half of the eighteenth. These would inevitably have led to changes in organ design [7], and thus to differences leading to the two styles of organ building which emerged in northern and central Germany. The situation resulted in the appearance of a distinctly 'Thuringian' type of small instrument of which that at Arnstadt was an example. Even so, evidence is easily found suggesting that the Arnstadt organ was perhaps a somewhat extreme instance of the breed nevertheless. One can look forty years later at Gottfried Silbermann's village organ at Fraureuth in Saxony, some 130 km (80 miles) to the east, which still retained the complete manual chorus work that also characterised the late seventeenth century northern school of organ building (Table 3). The only major point of convergence between the Arnstadt and Fraureuth dispositions lies in their emasculated pedal organs and their consequent reliance on a coupler to the chief manual. Otherwise they lie poles apart, just as the Schnitger and Wender organs did.

 

Hauptwerk 

 

 

Oberwerk 

 

 

Pedal 

Prinzipal

  8

 

Gedackt

  8

 

Subbass

16

Rohrflöte

  8

 

Rohrflöte

  4

 

Posaune

16

Quintadena

  8

 

Nasat

2 2/3

 

Oktave

  8

Oktave

  4

 

Oktave

  2

 

Spitzflöte

  4 

 

Quinte 

  1 1/3

 

Hw - Ped
Quinte

  2 2/3

 

Sifflöte

  1 

 

Superoktave

  2 

 

Sesquialtera

 II 

 

Tierce

  1 3/5

 

Zimbel

 II 

 

Mixtur

 IV 

 

 

 

Cornet III - V
Tremulant
   
Ow - Hw

 

Table 3. Dorfkirche, Fraureuth (Gottfried Silbermann, 1742)

 

One only has to listen to today's reconstructed Arnstadt instrument for a while before the remarks above, mainly concerning the under-developed chorus work, confront one forcefully. The performer can either use the wide range of individual timbres, or a few combinations such as 8 foot plus 4 foot, or add the mixtures. In other words, the organ is either quiet, mezzo-forte, or it becomes suddenly loud. There are not many dynamic options lying between the quietness and the loudness. It is a most curious situation, indeed I find its limited combinational palette rather tiresome in terms of aural experience, and it makes one long for clues as to how Bach would have treated the instrument. Looking to the usual authorities to provide the clues seems pretty fruitless in view of their widespread uncertainties over such basic details as to what the original stop list might have been, as noted earlier. Of course, we cannot know for sure what the original Arnstadt organ sounded like nor how it might have been played, but recordings are widely available of Hoffmann's carefully reconstructed instrument (e.g. [2]). Some YouTube demonstrations of the organ are also worth listening to (e.g. [9]).

 

 

The Hesse organ

 

So we have set the scene for Bach's musical workspace at Arnstadt. He would probably have had the whole organ gallery to himself - who else beyond his singers, his instrumentalists and the organ blowers would have traipsed up and down all those stairs? However it would have been a rather dark and forbidding place, since the dormer windows visible in the modern picture above were not inserted until the 1730s, and by then Bach had long since departed. The murkiness would have implied the need for much candlepower close to a lot of music paper and woodwork, not only expensive but perhaps causing some anxiety given that the previous church had burnt down.

Subsequent to the new windows, further redecoration and reordering of the church took place in the 1770s, a date which is significant since the building was taken backwards to this time when the Hoffmann instrument was installed in 1999. Wender's organ itself seems to have survived pretty much intact for over 150 years until repairs and major modifications were documented in church records in the 1860s [3], though there must surely have been other interventions over such a long period just to keep it playable. The 1860s work was instigated by the organist Heinrich Stade and executed by Julius Hesse, though it was apparently not an overwhelming success, partly because the instrument was enlarged to 59 stops across three manuals and it emerged with an unpleasantly heavy touch [10]. Nevertheless the newly conflated instrument attracted visits from Liszt on more than one occasion. One wonders rhetorically why the heavily attenuated connection to Bach in the mutilated organ apparently remained so magnetic to him.

 

 

The Steinmeyer organ

 

Any remaining Bach connection became even more tenuous when Steinmeyer was called in to build a new organ in 1913. This essentially romantic instrument had tubular pneumatic action with three manuals and 55 stops, and at first it was installed on the second gallery below that where Wender's organ had stood. In fact the third gallery was demolished to make room for the imposing case of the new instrument. However six old stops were retained, labelled as 'Bach Registers', together with parts of the old casework. This organ was subsequently lowered still further to a position on the first gallery in 1938 when the second was removed in its turn to make way for it. Not to put too fine a point on it, not only had Wender's organ been destroyed but the church was disfigured as well.

 

 

The Hoffmann organ

 

In 1997 Orgelbau Hoffmann [11] began a comprehensive reconstruction, though perhaps recreation is a better word, of the original Wender instrument. About 25% of the original pipes, 320 in total, were recovered from the Steinmeyer organ. This allowed four stops (Viola di Gamba 8, Grob gedackt 8, Octava 4 and Still gedackt 8) to regain almost all of their earlier pipework, together with three others (Gemshorn 8, Nachthorn 4 and Quinta dena 8) which also incorporate substantial amounts of old material. In contrast, four stops (Principal 8, Principal 4, Quinta 6 and Mixtur III) each only retain a single old pipe. No original pipes from the pedal organ were available.

Meticulous measurements were made of the old console in the Arnstadt Bach museum to enable it to be recreated down to the last detail. Evidence from other Wender organs made plausible realisations possible of the winding system (using four wedge bellows), wind trunks, wind chests, action, and an original Sub Bass stop. Three new reed stops were made using information from a 1728 Herbst organ at Lahm and a 1708 Contius instrument at Abbenrode. Of these, the two larger ones (Trompet 8 and Posaunen Bass 16) used fractional length resonators for the larger pipes because of the limited height available above the organ as mentioned earlier. Scales of some missing pipework were estimated from extant rack boards, and metallurgical analysis enabled the new pipes to be made from identical alloys to the old.

The arpeggiated sounds of the two Cymbalsterne are produced by tuned gongs similar to large bicycle bells. These are struck by small hammers lifted by rudimentary cams affixed to rotating shafts. The latter also project through two pipe flats in the casework to actuate the obligatory external 'stars'. Motive power for the whole affair is provided by two windmill-like turbines within the organ driving the internal ends of the shaftwork. Apart from the gongs and various small parts, the whole of this mechanism is of wood.

Working drawings for the project were based on the 'Weimar Foot' length unit (282 mm).

With so much new pipework, scaled, made and voiced from scratch, there is inevitably more doubt than certainty as to how closely the sound of the organ approaches that which Bach would have known, and this is a question which will never be resolved. Even the old pipework only constitutes about 25% of the total, and what Steinmeyer might have done to it is anybody's guess. To take just one example, if he had increased the mouth cut-ups of some pipes their speech would have been altered once and for all because it is next to impossible to reverse such an operation for obvious reasons, and because there is now no knowledge of what the original cut-ups would have been in any case.  As mentioned already, recordings of the organ are available (e.g. [2], [9]).

The new instrument was tuned to A465 at a temperature of 18 Celsius and at a wind pressure of 72 mm of water (2.8 inches). This is a semitone above today's standard pitch (A440). According to the organ builders [3] the temperament was set to 'Wender's unequal temperament' comprising seven pure and five tempered fifths. This choice was apparently informed by a letter from Johann Kuhnau (possibly to Johann Mattheson) in 1717 which said that Wender did not use an 'exact Neidhardt temperament'. However the two statements are only compatible if we suggest that the Wender tuning in fact comprised five pure and seven tempered fifths, i.e. the reverse of that stated above. This is because the Neidhardt temperament uses four pure and eight tempered fifths. So it is possible there is a misprint in reference [3] here. A further difficulty is that Neidhardt's temperament dates from 1724, seven years after the letter was apparently dated. Nevertheless, Neidhardt's is a very good temperament in which no keys are badly out of tune with the possible exception of E major. But it is well to restate that the original organ was built before he had developed it.

Hoffmann also rebuilt the Steinmeyer organ, though its visual impact in the church has now been reduced to zero. As can be seen from the photograph at the head of this article, it is entirely concealed behind featureless grilles underneath the Wender organ and it is played by electric action from a detached console on the first gallery. One speculates with interest as to which of the instruments is the most used today, however.

Running in parallel with the rebuilding of the organs was an equally ambitious restoration of the church. Obviously the third gallery, removed in 1913 to make way for the Steinmeyer organ, had to be reinstated to take the new instrument. As to the remainder of the interior of the building, it was decided to return it to its well-documented state after the work carried out in the 1770s which was mentioned earlier. The temptation to return it to how Bach would have seen it in 1703 was resisted, partly because there is incomplete knowledge of the internal layout at that time and partly because the necessary removal of the high dormer windows would have rendered the church (and particularly the remote third gallery with its new organ) rather dark and gloomy once again.

 

 

Concluding remarks

 

It has been said that "Bach played German organs of every type from late renaissance vintage with 'spring chests', to the mature works of Gottfried Silbermann in which the seeds of romanticism were already sown" [12]. If this was true of Silbermann's organs then it was writ larger in Wender's instrument of 1703 at Arnstadt, even though it predated Silbermann's later work by some forty years. Its design also supports Wills who argued that, in contrast to the seventeenth century Werkprinzip instrument of northern Germany, organists were combining principals, flutes and strings for bold effects rather than building vertical choruses in the old manner based on the natural harmonic series of pipe sounds [7]. At Arnstadt and in similar instruments around that time, mixtures were being retained more to boost loudness when necessary, rather than being integrated seamlessly into an holistic whole as previously. Players were also thickening solo lines with flutes and reeds at pitches no higher than 4 foot. The Arnstadt organ would certainly have encouraged and enabled them to do those things.

The various threads drawn together in this article support this. Indeed, if they did not, then one could be forgiven for wondering whether the tonal design of the Arnstadt organ defies logic. Its vertical choruses were incomplete in terms of pitch, incorporating deliberate gaps between the mixtures and the fluework below. There were almost no possibilities for synthetic tone building. The curious sub-quint mutation stop at 5 1/3 foot pitch was properly part of the 16 foot harmonic series, yet there was no 16 foot tone on the manuals. Was Wender deluded by the contemporary progress in musical acoustics to believe, mistakenly, that resultant tones with the unisons would emerge to do the same job? If so, he was far from alone since this deep-rooted misconception about 'difference tones' persists to this day among some musicians and organ builders [13]. If he had wanted real 16 foot tone in an organ where the required height was unavailable, could he not have used a stopped unison rank? The emasculated pedal organ at Arnstadt, also employed widely by Silbermann and others, foreshadowed the pitiful efforts so ubiquitous on small British Victorian organs 150 years later.

If there is logic in the design of the Arnstadt organ, perhaps it confirms that people were getting bored with the restless contrapuntal polyphony they had endured for so long, and that they wanted to relax instead with more melody, lyrical textures, individual colours and combinational subtleties driven by timbre rather than pitch. Did Bach like the instrument? Who can say? He only stayed for four years despite having a brand new organ under his hands while still in his teens. All we know is that C P E Bach said of his late father that he never had the opportunity to preside at a first rate organ. If that were true, then perhaps his poignant remark says it all.

 

 

Notes and references

 

1. "Are we any closer to understanding J S Bach, the organist and organ-composer?", Peter Williams, The IAO Millennium Book, Incorporated Association of Organists, July 2000.

2. "Johann Sebastian Bach in Arnstadt", Gottfried Preller spielt an der Wender-Hoffmann-Orgel Werke von Johann Sebastian Bach.

An audio CD produced in 2000 by Orgelbau Hoffmann, Ostheim/Rhön and Architekturbüro Müller GbR, Blankenhain/Thüringen.

3. The value of the CD mentioned in [2] above is twofold as far as this article is concerned. Besides enabling the reconstructed organ to be heard as played by the Director of Music at the church, it is accompanied by an unusually comprehensive, well-produced and illustrated sleeve note. This 23 page booklet (in German) is more of a research report than a mere sleeve note. It was compiled by the architects and organ builders responsible for the reconstruction of the building and its organ which took place between 1997 and 2000. It does not appear to be available independently however.

4. "The Organ", Peter Williams & Barbara Owen, Macmillan, London, 1988.

5. "The Organ", W L Sumner, Macdonald & Co, 3rd edition, London 1962.

6. "J S Bach", Albert Schweitzer 1908, trans. E Newman, facsimile edition Dover, New York, 1966.

7. "Organ", Arthur Wills, Kahn & Averill, London 1997.

8. "The physics of organ blowing", an article on this website, C E Pykett 2020.

9. "Demo of the 1703 Wender organ in Arnstadt by Balint Karosi"

See http://www.youtube.com/watch?v=wCzggf6bv9E

In view of the eminence and skill of the performer it is unfortunate that the sound track here is of uneven quality. It was probably made with mediocre equipment and the signals subjected to dynamic compression. This can sometimes be detected as the reverberant sound dies away, when there is a sudden jump in loudness towards the end of the reverb tail. This also means that the volumes of individual stops and combinations thereof probably do not reflect their actual relative loudnesses, in addition to the different recording levels which might have been used. Therefore the recording is mainly useful for comparing their tone qualities. Having said all this, one can nevertheless take away quite a lot from the recording.

10. Some anecdotes suggest that Hesse's work in the 1860s might not have been of the highest quality, and moreover that he vanished at some point having already pocketed some of the money.

11. The firm formerly known as Orgelbau Hoffmann in 1999 later became Hoffmann & Schindler, not to be confused with Heinz Hoffmann Orgelbau of Stelle.

12. "The Baroque Organ", W L Sumner, Proceedings of the Royal Musical Association, 11 November 1954.

13. "Resultant bass, beats and difference tones - the facts", an article on this website, C E Pykett 2011.