PART 3/4

Dealing with and interpreting historical sources.

(Type Ctrl+F to find any text on this page)

   THE YEAR 1657

In parts 1 and 2 we have identified the main characters, mentioned in the book Innovation & Collaboration - The early development of the pendulum clock in London (Garnier & Hollis)(1, based on historical facts. This third part deals with the year 1657 in which the invention of Christiaan Huygens, the application of the pendulum in a clock movement, was made into a working and commercial product. In addition, we will discuss and denounce the research method that, especially from England, has been used now for decades. It is striking, that this method always has a particular underlying strategy: to reduce the role of Salomon Coster, and to a lesser extent the role of Christiaan Huygens, to increase the role of the English makers and in particular that of John or Ahasuerus Fromanteel.

As an example of this method, we want to examine the new theory of Garnier & Hollis, based on facts, as documented in the archives. This theory was widely promoted during the beautiful exhibition Innovation & Collaboration, held at Bonhams in London from 3 to 14 September 2018 and also recorded in the aforementioned book.
The surprising conclusion of Garnier & Hollis is that John Fromanteel came from London to The Hague to teach Salomon Coster, between September 1657 and May 1658, how to make a pendulum clock.


Before we go into the story of Garnier & Hollis first the facts at a glance: the patent for the application of the pendulum in a clock movement was granted by the States General to Salomon Coster on 16 June 1657.(2 With this, Coster had the (exclusive) right for 21 years to make pendulum regulated clocks in Holland and West-Friesland. Unfortunately, the patent application itself has not yet been found or has been lost altogether, but it obviously had been submitted some time before the conferment on 16 June.

This implies there was only a period of at most a few months between the actual invention of Christiaan Huygens (December 25, 1656) and the conferment of the patent. In the experimental phase from January 1657 to the patent application, work will have been done to make a functioning pendulum clock. Although nowhere has been recorded who assisted Huygens during this first phase, the most likely candidate is Salomon Coster. After all, he was the applicant for the patent. Also at a later stage (December 1657), during the works in the church tower of Scheveningen, Coster was obviously the right man to call upon when it involved experiments with pendulum movements.
(3 No other clockmakers assisting Huygens are mentioned in this first period.


Salomon Coster was at that time one of the most important clock- and watchmakers in the Netherlands, having made complicated box clocks and watches. Given the short time span between the simple but brilliant invention of Huygens and the application of the patent, it was apparently easy for Coster to construct four gearwheels between two plates, a pendulum suspension and motion work. This new construction was much simpler than the movements Coster had made so far.

After obtaining the patent in June 1657, Coster will soon have found out that he had opened a new market and that the expansion of labour capacity was desperately needed. He will have sought cooperation in order to enlarge the current production and to be able to serve the new market of the pendulum clock.

As an advanced apprentice John Fromanteel fitted well in this picture. He was a 5th-year apprentice, not yet a journeyman, so not yet an accomplished clockmaker. On the other hand Fromanteel was not a starting inexperienced student, still taking up a lot of time from Coster to teach all the principles of clockmaking, but immediately a productive employee.

John was hired and in September 1657 an employment agreement was drawn up for the duration of 8 to 9 months. This agreement, nowadays also known as the Coster-Fromanteel contract, has already for decades been a source of the most peculiar theories.(4

When we look at the facts in the contract, we see little more than a fixed-term employment agreement with performance-based pay, between master clockmaker Salomon Coster and skilled apprentice John Fromanteel.


Fig. 1
The contract

These ‘job titles’ are historically correct, since Coster has been named as such since 1645 and John Fromanteel, in the fifth year of his apprenticeship, was still an advanced student, also according to the rules of the Worshipful Company of Clockmakers. Moreover, these job titles are exactly defined in the employment agreement between Coster and Fromanteel. The work of skilled apprentice Fromanteel takes place in the workshop of the master clockmaker Coster and a pay rate per piece is agreed. On the basis of either the currently still existent Coster pendulum clocks, or archival documents, it is not possible to determine who did what in the production process. Even the agreement between Coster and Fromanteel only mentions the general words "werck" and "orlogiewerck" (i.e. movement or clock/watch movement), which implicates that any type of movement could be meant. However, it is becoming increasingly clear that the Coster workshop had a much larger production than was thought for a long time, and that also several other clockmakers were working for Coster.(5

Many stories have been written and thoughts expressed zooming in on the smallest parts (even up till collet level!). Just the same, we will stay away from these micro level discussions, because evidence on who did what exactly, can never be delivered.

It would be better to have the metal and wood of the clocks scientifically examined by experts using the most modern equipment.

If scientifically proven authentic, the fact that these are Salomon Coster clocks is evident, just because the clocks are signed with his name, just as The Nightwatch was signed by Rembrandt, being a collaboration of several painters under the wings of the master. Or a Fromanteel signed clock, on which several clockmakers will have participated.

Other matters in the agreement, such as the supply of materials, explaining single phrases, points and commas in the text, can all be interpreted in various ways, but without additional irrefutable evidence, they are good for a rainy afternoon discussion, but scientifically uninteresting.

One good example of the lack of historical knowledge and thorough research in these discussions is the fact that by some authors a lot of significance and various explanations are attributed to the numerous erasures in the agreement between Coster and Fromanteel.

fig. 2
Same notary book more erasures

If these authors would simply study the same notary book, or other 17th century Dutch notarial protocols, they would see that many deeds contain the same number or even more erasures (Fig. 2). Unfortunately, this is conveniently overlooked here.

Nevertheless, in the absence of any important and unambiguous English archives about the first pendulum clocks, especially our English clock friends for decades now are trying to find evidence in this contract in favour of Fromanteel in particular.


Whereas previous English speculations assumed an important role for John Fromanteel in the manufacture of the first pendulum clocks, Garnier added a new chapter to the Coster-Fromanteel discussion in the book Innovation & Collaboration. In chapter 3, Garnier & Hollis try to build a story in which is stated that apprentice John Fromanteel came to The Hague to explain to master clockmaker Coster, in eight to nine months, how to make a pendulum clock. At critical moments, where the reader expects an important archive piece or other conclusive evidence, remarkably often the words "suggest", "maybe" and "strongly possible" are used.

Although we stick to archival documents as much as possible, it sometimes is unavoidable to propose a highly probable theory. Be that as it may, this theory then must be firmly supported by facts. When a theory aims to demonstrate a major change in historiography, the evidence must be strong and conclusive.

As an example: in Tijdschrift 18/2(6 it is said that the brother of Jannetje Hartloop (Coster's wife) presumably learned the trade at Coster’s. The proof for this is not conclusive, but it is plausible. Moreover, and this is important, it is a theory that does not turn history upside down.

When the attribution of the first pendulum clocks and the role during the experimental phase is moved from Coster to Fromanteel, this is of a different order and the evidence must be strong and supported by multiple archival records.

In their book, Garnier & Hollis have proclaimed the remarkable theory that Fromanteel taught the making of a pendulum clock to Coster, instead of the other way around. For this conclusion they use the following structure:

1.      The weight driven clocks with seconds indication were the basis of the patent application. Thuret(I was the one who assisted Huygens in this experimental period and in the first quarter of 1657 was responsible for making the pendulum clocks driven by weights.

2.      In Horologium (1658) Huygens speaks about 'manufacturers'. This word is in the plural, so if Thuret assisted Huygens, there must have been another clockmaker assisting Huygens in this experimental phase. Who was this other clockmaker?

3.      The Coster-Fromanteel agreement has a sensational phrase that shows that Fromanteel demonstrates the making of a pendulum clock to Coster, instead of the other way around.

4.      In addition to Thuret, Fromanteel therefore was the other one assisting Huygens in the experimental phase of the pendulum clock. Where Thuret was responsible for the weight driven clocks, Fromanteel was responsible for the ‘commercial’ spring driven movements.


In the run-up to Garnier & Hollis' new discovery in the Coster-Fromanteel agreement (more on this later in this article) they immediately make a huge mistake. They state that we should realise that a weight driven clock with a seconds indication was the basis for the patent application. The main shortcoming is, the patent application has so far not been found in the archives and the grant of the patent itself does not mention a weight clock.

End of this section, click here to continue.


I Isaac Thuret ca. 1630 – 1706), horloger ordinaire du roi, was one of the most productive makers in Paris in the 2nd half of the 17th century. He was also the clockmaker responsible for the maintenance of the machines of the Académy des Sciences and the Paris observatory. In 1675, Christiaan Huygens asked the help of Thuret to produce the first spiral spring watch. In January 1686, Thuret moved into the ‘Galleries du Louvres’.
II  ‘Ingevolge de bevoegdheid in het Koninklijk Besluit van 23 augustus 1907 Staatsblad no. 237), gegeven aan gemeenten, die aan zekere voorwaarden voldoen, om de archieven van de notarissen, die hun standplaats binnen die gemeenten gehad hebben, van het Rijk in bewaring te ontvangen, heeft de gemeente 's-Gravenhage in november 1910 alle archieven van op haar toenmalig grondgebied geresideerd hebbende notarissen over de jaren 1597-1811 op haar verzoek in bruikleen ontvangen.’

 ‘sal indemneeren ende vrijhouden van bier, vuur ende licht’

1  Richard Garnier & Leo Hollis, Innovation & Collaboration, The early development of the pendulum clock in London, Fromanteel Ltd., Isle of Man, 2018
2  Nationaal Archief, tnr., Staten van Holland, 1611, nr. 22, d.d. 16-07-1657.
3  O.a. in Oeuvres Complètes de Christiaan Huygens, Tome Deuxième, correspondance 1657-1659. p. 108. Nr. 433.
4  Haags Gemeentearchief, tnr. 0372-01, Notarieel archief Den Haag, 322, folio 409.
5  Although it was drafted two years later, the estate statement, drawn up immediately after Salomon Costers' death, shows that there was a huge stock in the Coster workshop and that Coster also outsourced work to Severijn Oosterwijck. See: Victor Kersing and Rob Memel, ‘In de voetsporen van Salomon Coster. Van Hagestraat naar Wagenstraat’ in: TIJDschrift 18/2. p. 4-9.
6  Kersing en Memel, ‘In de voetsporen van Salomon Coster Van Hagestraat naar Wagenstraat’ in: TIJDschrift 18/2. p. 4-9.
7  Translation from Tijdschrift voor horlogemakers, 1 maart 1903.
8  O.a. in Going Dutch. The invention of the pendulum clock. Proceedings of a symposium at Teylers Museum, Haarlem, 3 december 2011 Zaandam, 2013). p. 29.
9  For more on this topic and Thuret see part 4 of these articles: Part 4: The invention of the pendulum clock – The Sequel, more inventions.
10  Haags Gemeentearchief, tnr. 0372-01, Notarieel archief Den Haag, 322, folio 409.
11  Richard Garnier & Leo Hollis, Innovation & Collaboration, The early development of the pendulum clock in London, Fromanteel Ltd., Isle of Man, 2018, pp. 67
12  Richard Garnier & Leo Hollis, Innovation & Collaboration, The early development of the pendulum clock in London, Fromanteel Ltd., Isle of Man, 2018, pp. 68
13  John C. Taylor, The Salomon Coster John Fromanteel contract 3rd september 1657. Changes made during drafting s.l., s.a.).
14  John C. Taylor, The Salomon Coster John Fromanteel contract 3rd september 1657. Changes made during drafting s.l., s.a.), pp. 2 point E.
15  H. Brugmans, ‘Uit de protocollen der Haagsche notarissen’ in: Die Haghe. Bijdragen en mededeelingen 1908, p. 1-124; H. Brugmans, ‘Uit de protocollen der Haagsche notarissen II’ in: Die Haghe. Bijdragen en mededeelingen 1909, p. 1-78; H. Brugmans, ‘Uit de protocollen der Haagsche notarissen III’ in: Die Haghe. Bijdragen en mededeelingen 1910, p. 1-113.


Garnier & Hollis then draw a wrong conclusion from Horologium (September 1658)(Fig. 3). Huygens writes in his concluding speech about the uniformity and the firmness of the pendulum.


Fig 3. Horologium

The translation(7 from Latin is as follows:

Much that I could add to this, I leave to the ingenuity of the manufacturers, who, once they have understood my invention, can easily find out how it can be applied to the different types of movements and also to those that have been made according to the old system’.

Huygens concludes his Horologium with the remark that his invention will be further developed and that manufacturers (among others clockmakers) will accomplish this.

Huygens also states, with the then current insight, the existing balance/foliot movements are easily converted to a pendulum. Huygens made this statement more than a year and a half after his invention, after many pendulum clocks had already been produced. Garnier & Hollis mistakenly see this as the evidence that several people were involved during the experimental phase of the pendulum clock (early 1657).

Here the story can actually stop, but we will take you into some other curious next steps.

Reference is made to a publication by Sebastian Whitestone stating that Thuret was the clockmaker helping Huygens with the experimental pendulum clocks in early 1657.

However, Thuret only comes into the picture in the Oeuvres Complètes of Huygens from 1662 onwards. Up till now, no document has shown Huygens knew Thuret before this date, let alone they were working together during the early phase of the pendulum clock.

Probably, the confusion arose because the editor of Oeuvres Complètes, suggests in the margin of a letter from Huygens to Cha<>pelain of 20 August 1659, , Thuret may be meant by the word 'he'.
Unfortunately, the Oeuvres Complètes do not show the basis for this suggestion. Nor have we been able to find anything in this regard not show the basis for this suggestion. Nor have we been able to find anything in this regard elsewhere.

We do know some still existing wall clocks with long pendulum signed by Thuret.(9

On the basis of various characteristics of both case and movement, these clocks need to be dated much later than 1657.

Thuret long pendulum wall clock

Fig. 4 
Much later than 1657.
Weight driven long pendulum
wall clock by Isaac Thuret á Paris

Garnier & Hollis then return to September 1657 to explain the new major discovery by quoting the Coster-Fromanteel agreement.
The following sentence appears in the Coster-Fromanteel contract

soo heeft hij Coster belooft hetselve werck dat hij Fromanteel sal maecken I (ende het secreet daerinne bestaende), hem voor den voorsz. geexpereerden tijt te openbaeren, mits dat de wercken die bij hem Fromanteel hem op conditien hiervooren verhaelt sullen sijn gemaeckt hij Coster voor de voorbedongen prijs aen sich belooft sal mogen ende moeten behouden.(10

This is translated to:

... Further so has he Coster promised to reveal the same work that he Fromanteel will make (and the secret therein existing) to him before the aforementioned expired time, provided

that the works which by him Fromanteel on conditions aforementioned will have been made, he Coster for the stipulated price will be allowed and obliged to keep.(11

Or in plain English: Coster will tell Fromanteel the secret that is in the (pendulum) movement and Coster will also take the manufactured movements for the pre-agreed price.

The handwriting of notary Josua de Putter is not always clear and his commas and other punctuation marks sometimes differ. However, Garnier & Hollis see two punctuation marks as extra parenthesis and the text according to Garnier & Hollis gets a completely different meaning (Fig. 5). The new sentence now becomes (see new brackets) as follows:

soo heeft hij Coster belooft hetselve werck dat hij Fromanteel sal maecken ((ende het secreet daerinne bestaende), hem voor den voorsz. geexpereerden tijt te openbaeren) mits dat de wercken die bij hem Fromanteel op conditien hiervooren verhaelt sullen sijn gemaeckt hij Coster voor de voorbedongen prijs aen sich belooft sal mogen ende moeten behouden.

Which then suddenly is translated to:

... furthermore during the aforementioned time so has he Coster contracted the same work that he Fromanteel will make - (and the secret incorporated therein) him to explain before

the aforementioned expired time - provided that the works which by him Fromanteel on conditions previously cited shall be made the same he Coster for the pre-stipulated price to
himself promised will be allowed and obliged to keep...(12

So we see a larger part of the sentence in brackets. In this Garnier & Hollis see proof that Coster did not tell the secret to Fromanteel, but the other way around: Fromanteel taught Coster how to make a pendulum clock!

Fig. 5

We, and various experts, once again have looked at the meaning of this new sentence structure and into the scope of the entire agreement. Unanimously we come to the conclusion that this new theory, in which punctuation marks are seen as extra brackets, (fig. 5) does not result in any change in the meaning of the text. In no way it is comprehensible or possible to follow the conclusions of Garnier & Hollis.


The tales on the new Fromanteel story are based on far-fetched assumptions and theories. It actually entails a recommendation to investigate in the future on the basis of facts laid down in dated deeds and to take note of the 17th century Dutch manuscripts, the use of language and the course of events at 17th century notaries.

Of course one can put forward ideas and thoughts at certain moments, but when the authors think to have made great discoveries, as is the case with Innovation & Collaboration, they will have to come better prepared than bring a theory built on quicksand.

  TWO G's

On the occasion of the exhibition Innovation & Collaboration, Dr. John C. Taylor published and commented a facsimile edition of the ‘Coster-Fromanteel contract.(13

Taylor Coster Fromanteel Contract

The facsimile looks beautiful and well cared for; clearly, a lot of attention has been paid to this, as well as to the contents of the contract itself. The document has no less than 67 notes with comments. This enormous number alone already suggests the author is not accustomed to reading seventeenth-century Dutch notarial protocols, let alone interpreting them. This is immediately apparent for paragraph 1]. The author sees two lines in the top left hand corner, drawn through the original text. That's right, but they are not just lines or erasures. This are two G's indicating two copies, two neat copies, have been made of this deed, which may or may not have been certified by the notary.

fig. 6  Two G's in the top left corner


fig. 7  One G in the top left corner


This clarifies point I of the introduction. Here the author states, according to the authority of Prof. Lisa Jardin, that no legal system would have such a sloppy document filed.
However, this was common in 17th century Holland.
And where were they archived? At the Court of Holland. After all, this body admitted the notaries in Holland. So the archives were in custody with the State. And how did they end up at the municipality of The Hague? We quote from the introduction of the inventory of the notarial archive in The Hague:

‘Pursuant to the authority from the Royal Decree of 23 August 1907 (Bulletin of Acts and Decrees no. 237), given to municipalities, that meet certain conditions, to keep the archives of the notaries, who have had their place of employment within those municipalities, from the central government in custody, on November 1910, the municipality of The Hague received all archives of notaries who had resided at its territory in the years 1597-1811 at its request on loan.’ (II

This is just an aside, for a better understanding of the author, who asks himself in note 9 ‘How this contract can be viewed as draft minutes and where the protocol is stored remains unclear’. On the contrary, that is as clear as daylight.
Besides, in contrast to nowadays, notarial deeds were not always drawn up at a chic notary office. Also changes in the text were not initialled as they are today. A lot of these procedures even happened in the pub. The world really looked a lot different in the 17th century.


Next to mistakes due to lack of historical knowledge, there are also reading errors. The author is forced to make use of an English translation, where the danger is that there are differences in nuance or even differences in meaning. An example: in point K of the introduction, the author says:

The contract does not spell out normal issues which are taken for granted; for example, the cost of John’s travel to The Hague is paid by John: Salomon will provide John’s board and lodging. So why are beer, fire and light mentioned if they were normally supplied by masters to their workers?

Indeed, the agreement does not include things that are normal. But what was normal back then? That really was not fixed. Subsequently the author claims that Salomon would provide John ‘board and lodgings’. But that is not what it says. It does only say that Coster will

indemnify and defray Fromanteel (usually to put synonyms in succession with beer, fire and light

so he will provide him with beer, heating and light to be able to do his work. This also means Coster does not provide him with accommodations and meals. So, Fromanteel stays/sleeps elsewhere.


But there is more, maybe even more serious. Let us again limit ourselves to some examples, because there are a lot of mistakes and there is no room to cover everything.
Dr. John C. Taylor, an entrepreneur and inventor, who has registered more than 400 patents worldwide, states he has never yet lost a lawsuit about a patent worldwide. Truly a formidable achievement that certainly commands respect. But this look-and-see statement suggests that he also has knowledge of patent law and contracts in 17th century Holland.
This is nothing more than a suggestion, because apparently he does not have that knowledge. He looks at 17th century Holland through biased 20th and 21st century international glasses.
An important example of this is the introduction of the necessary existence of a Heads of Agreement.

Parties coming for a lawyer usually have at the very least a Heads of Agreement worked out between them, together with a wish to compromise to some degree to achieve a final Agreement

That may be customary nowadays, but was this so in the 17th century? The author states it is very common - ‘at the very least’ - so he does assume it was so at the time. In this way he’s working towards a larger role of Fromanteel.
What did the authors want to achieve with their entire exercise? They wanted to show that a normal agreement of employment between Coster and Fromanteel is in fact a learning agreement between Fromanteel and Coster, i.e. a 180 degrees rotation of roles, together with a change in the value of the agreement itself. In our opinion the authors have not succeeded. Too few arguments have been put forward for this radical change, and at the same time the authors have disqualified themselves by showing that they have no knowledge of 17th century notarial practice in Holland.

Prof. Dr. H. BrugmansAs they are unable to read archival documents, they for instance would have done themselves and all readers a favour to have first been informed by Prof. Dr. Brugmans, Professor of General History. He more than a century ago published a collection of 180 work and student agreements from the Hague protocols. (15

These very quickly show that the world then did not look as unambiguous and tightly organised as the authors suggest.


March 2019, Copyright:

(This article is subject to ongoing revisions.)


Chr. Huygens' Œuvres Complètes. (pdf)

Coster Fromanteel contract.

Archive research on Salomon Coster.

Chr. Huygens Horologium 1658. (pdf)