Legal Expropriation

I. A Puzzle for Specialists

The principle is so simple that every child knows it from the Spirograph. Workpiece and tool rotate at the same time, their axes are offset from one another, and the shape arises from the ratio of their speeds: a triangle, a hexagon, a polygon, inside as well as outside, in a single clamping, without reciprocating masses. In 1982 Hans Ley filed a patent application for the process and optimised it with further applications.

The patents were granted and it worked. Heckler & Koch announced the process at the AMB in Stuttgart in 1988 with a first brochure and presented the “Formdrehzentrum FDZ 100 System Ley” (form-turning centre) at the METAV in Düsseldorf in 1990. In 1990 this machine was used to turn the receivers and shanks of the Capto tool-clamping system made by the Swedish company Sandvik, which has been standardised worldwide as PSC in ISO 26623 since 2008. The cage rings of a large rolling-bearing manufacturer, the Swedish company SKF, have been produced in series according to this principle for more than thirty years.

Polygon shaft-hub connections, which the invention also makes it possible to produce economically, transmit more torque in a smaller space than the feather key or the spline and serrated shafts that remain standard in mechanical engineering to this day. Polygon connections used to be regarded as better but too expensive to manufacture. Since the inventor presented his process at the EMO in Hanover in 1985, that no longer holds: since then, polygon connections have been not only better but also more economical to produce than the usual types of connection. The most striking figure for this was supplied, of all people, by Prof. Peter Werner Gold of RWTH Aachen in 2006, in his article “In acht Sekunden zum Polygon” (“To the Polygon in Eight Seconds”): roughly eight and a half seconds for a polygon machined from solid stock, roughing and finishing included. He did not mention the inventor’s name.

Every unprejudiced specialist who sees this asks the same question: why has so simple and so useful an invention not found a market commensurate with it in forty years? The answer does not lie in the technology. It lies in the conditions under which an invention has to become an innovation, and in the people who stand at those conditions.

II. The Typical Situation of the Inventor

An independent inventor has something that many others lack: the knowledge, the idea, the understanding of the principle. He lacks what a few others have: capital, manufacturing, distribution, access to customers. The typical independent inventor is a have without means, the counterpart to Günther Anders’ have-not with means (“bemittelter Habenichts”).

From this imbalance follows an asymmetry of time. Developing and building machines costs money, patent attorneys are expensive, patents cost vast sums, of which the annual fees are the least, and patents expire after twenty years. The inventor has to live, finance his development, defend his rights. Those facing him — companies, banks, institutes — can wait. Every year without a reasonable agreement on equal terms weakens the inventor and strengthens the others. In the end he does not need to be persuaded. He is exhausted.

The pattern is not confined to technology. The songwriter who signs his rights over to a label because he has to pay the rent. The farmer who sells his harvest at any price to the only wholesaler in the region because otherwise it will rot. The author who gives away his first novel for a fee he can live on for half a year. In every case it is not the value of the thing that decides the price, but the question of who can wait longer.

III. Chicken and Egg

The second condition is a chain in which everyone is waiting for everyone else.

A machine manufacturer does not build a machine for a process for which there is no demand. A designer does not use a polygon connection as long as there is no standard, no calculation rule and no machine that produces it. A standard comes into being only when enough users call for it. A calculation rule comes into being only when someone pays for the research, and research follows the interest of industry, which as yet has none.

Every link in this chain acts reasonably. Together they block one another. The standard for the geometry of the H-profiles appeared in 2021, almost forty years after the patent. The parts on calculation and manufacture are still outstanding. That is no sign of unsuitability; it is the speed at which a chicken-and-egg chain dissolves of its own accord.

IV. The Valley of Death

Between invention and market lies what innovation researchers call the valley of death: the phase in which a process has been proven technically but not yet tested economically. Research no longer finances this phase, the market does not yet. Most inventions die in this valley — not the bad ones, but those without a bridge.

For polygon turning there was such a bridge in the mid-nineties, after the Heckler & Koch bridge had collapsed in the peace crisis following the end of the Cold War and the subsequent bridges via MAHO and WIHA/INNOVAT had not proved sound. Together with the Prof. Gold already mentioned, the inventor had made contact with Ford’s transmission plant in Düren and proposed using his invention to manufacture the parts for constant-velocity joints, which were needed in large quantities. As he remembers it, Ford intended to produce a considerable proportion of them with the process. The development costs had been approved in writing by the group headquarters in Dearborn. Then GKN, the established manufacturer of such joints, submitted an offer to Ford which the inventor still regards as a sham offer. Ford in Dearborn put the project on hold. Later GKN withdrew the offer.

GKN knew the process. As early as 1983 and 1984 the inventor had been in contact with the group through a subsidiary. In 2002 GKN was to appear once more, with an opposition against a patent that emerged from the project we are describing here.

When Ford stopped the project, the inventor gave up. That the stop would only be temporary he did not know at the time. After so many years in which every bridge had proved brittle, he lacked the strength for another attempt. An employee in a corporation may interpret that as personal failure: whoever gives up after only fourteen years of struggle is no great loss; he deserved to go under. So judges the one whose salary arrives at the end of the month whether his project succeeds or not.

V. How the Game Turned

Adolf Kochsiek was not an established machine builder. He had lost a former company of his own and ran a small design office which, among other things, designed agricultural machinery for the firm Frost in Petershagen-Friedewalde. For the project with Ford he was envisaged as designer, and only as designer – a service provider.

At the start he was a potential contractor of the inventor. The inventor had the process, the patents, the know-how, the contact with Ford and the written commitment from Dearborn. Kochsiek was to receive a contract as soon as the agreement with Ford had been concluded.

Then GKN submitted the offer already mentioned. Ford in Dearborn put the project on hold, and the inventor, after so many years of struggle, gave up in frustration. When GKN’s offer became obsolete, the responsible head of development at Ford, Steinmeier, held the strings.

He had the renewed approval from Dearborn, he decided whether and with whom the project would continue, and when it revived he continued it — not, however, with the inventor, with whom he no longer had any contact, but with the designer Kochsiek, who had kept in touch. Kochsiek seized the opportunity. Who first had the idea of edging the inventor out, the inventor does not know.

He was brought back in only when Kochsiek informed him that he no longer wanted a partnership with him; he wanted to buy the patents. The one who had hoped for a contract had become the buyer; the one who should have awarded it, the seller.

Nord/GI, the investment company of Nord/LB, had taken a silent partnership of 500,000 marks in the inventor’s company in 1985 and in return had secured a lien on all his patents. After eleven years its patience had long since run out and it wanted to close the case with a small purchase price. Because it held the lien, it could sell the patents just as it could sell the machine. All the inventor could still do was sign. In the summer of 1996 he did so. Iprotec GmbH, based in Friedewalde, took over the patents, the applications, the machine and the knowledge. A few days or weeks earlier, on 21 June 1996, Kochsiek had filed a patent application for his own variant of the process.

VI. Everyone Optimises for Themselves

One can tell this story as the story of an individual who exploited another’s weakness. That falls short, for Kochsiek could only do what his surroundings permitted and rewarded. Each of those involved had his reasons.

Steinmeier, the head of development at Ford, bore the risk of a new process while the established supplier GKN stood ready. A counter-offer, whether seriously meant or not, provided the reason to stop. And when the project revived, a designer with no rights and no claims of his own was a more convenient partner than an inventor who wanted a say in his process.

GKN protected its own process. An offer that holds up a project costs little if it can be withdrawn later.

Professor Peter Werner Gold, who knew the inventor from earlier collaboration and had made contact with Ford together with him, needed projects, industrial partners and doctoral theses. A new company, run by an apparently experienced designer – whereas the inventor said of himself that he was no designer –, with a proper registered office and with a machine and patents cleverly acquired for little money from NORD/GI, was a better partner for everyone involved than a single, exhausted and frustrated inventor without a company. At Gold’s institute at RWTH Aachen a doctoral thesis was written in 2000 that is based on the machine now owned by Iprotec and that passes over the inventor in silence just as the 2006 publication already mentioned does. Throughout, Gold and his doctoral student Gödecke describe the firm Iprotec as the developer of the process. In the thesis the turned profiles even received the name of the patent holder: PI profiles, the I standing for Iprotec.

Nord/GI wanted to close a case that had cost it eleven years. It held the lien on the patents, and patents can be sold; patience cannot.

And Kochsiek, a designer without manufacturing of his own, got what he would never have achieved by his own efforts: a company with a machine, a finished process and the patent rights. He also got a powerful customer with an interesting development project for which he had done nothing and which, to the inventor’s knowledge, brought him a first payment of one million marks on signature of the contract.

Each of these decisions can be justified from the point of view of the one who made it. That is precisely where the mechanism lies. An expropriation needs no thief if everyone optimises for themselves and the sum of the optimisations calculates the inventor out of his own invention. In a valley of death what decides is not who made the invention or who understands it best, but who is still standing last when the others give up.

VII. The Expropriation of Authorship

Rights can be sold; authorship cannot. In the patent register Hans Ley remained entered as inventor even after the transfer to Iprotec in 2000. In the specialist literature he disappeared nonetheless.

In 2006 Gold presented the process in the journal “antriebstechnik” as a development of Iprotec and traced its history back to Musyl and the Viennese firm Krause. He did not mention the inventor. Gold need not have known the later stages: the form-turning centre FDZ 100, which Heckler & Koch presented at the METAV in Düsseldorf in 1990, and the WIHA S42, which the inventor showed at the EMO in Hanover in 1994. But he witnessed the beginning at close quarters. Gold was managing director of the firm Reintjes in Hameln, which manufactured the parts for the non-circular turning unit with which the inventor presented his process at the EMO in Hanover in 1985. An article in Maschinenmarkt followed in 1986 and a contribution to the Jahrbuch für Optik und Feinmechanik in 1987. In 2012 the VDI nachrichten described Iprotec as the inventor. Technical papers from 2012 to 2020 cite Iprotec’s patent or a later patent by the Bulgarian Jordan Maximov for the “two-spindle turning process” and pass over the real inventor in silence.

Each citation takes over the previous one. No individual strikes out the name. It drops out because an independent inventor has no institute that produces doctorates about him, no press office and no conference. The sociologist Robert K. Merton called this the Matthew effect: unto every one that hath shall be given.

When a researcher had the missing attribution pointed out to him in 2024, he suspected the error lay with the patent office, which must have overlooked the earlier patent. The patent office had not overlooked it: Iprotec’s own patent specification names it. It was the specialist literature that had overlooked it.

VIII. The United Expropriators

One could close the story here as an ordinary case of market power. The unusual part comes afterwards, and from the inventor’s point of view it is the real scandal: those who appropriated the process made nothing of it.

Marx wrote that in the end the expropriators would be expropriated. Here something more banal happened: they failed at what they had taken over. The project for manufacturing constant-velocity joints with which everything had begun petered out, by all that is known. The institute in Aachen produced a doctoral thesis and a journal article, but no market. Iprotec paid the patent fees until the end of the term and itself admitted in 2012 that the solution received hardly any attention in practice because designers fell back on familiar processes. That is exactly the chicken-and-egg problem the inventor had wanted to solve and that the buyers did not solve.

Technically, in the inventor’s assessment, Kochsiek never got beyond the simple two-spindle process. His only invention of his own in this field is a U-axis in which the tool is not displaced radially but swivelled about a transverse axis and adjusted hydraulically. Iprotec filed it for a patent in 1997; in 2016 Kochsiek filed a modified version again as a tool spindle, and the examiner cited the earlier application against it. The inventor regards this as a detour with disadvantages. Swivelling changes the engagement conditions at the tool, and hydraulics on a rotating system is nothing to enthuse users in machine-tool building, where hydraulics have largely died out. Tellingly, this patent specification too names the basic patent of 1982 as known prior art. As far as is known, Iprotec today produces polygons as contract work. The process that could have changed a whole class of machine elements has become a niche business.

From the inventor’s point of view this is a concerted incompetence. Each of those involved understood enough to secure the process for himself, but none enough to establish it. The institutions that supported them did not restrain them in this but encouraged them, for they reward the securing of rights, not the creation of a market.

The unprejudiced specialist therefore does not understand why the process found no market: he examines the technology and finds it good. But the obstacle lies not in the machine but in the machinery around it. Lewis Mumford called it the megamachine. It needs inventions, but not inventors. It absorbs the knowledge and casts off the person who produced it. And the damage strikes not only him but everyone who for thirty years has been using feather keys where a better connection would be possible. An expropriation without innovation is a loss without winners.

IX. What Would Refute This Essay

The thesis would be refuted if it could be shown that the process is technically or economically inferior and has therefore rightly found no market. Thirty years of series production at SKF and the research findings on the load capacity of the H-profiles speak against this. It would be weakened if the buyers had demonstrably made serious efforts to create a market and had failed because of external circumstances. And it would be overstated if authorship were correctly reported in the rest of the specialist literature. A research report of the Karlsruhe Institute of Technology from 2024 explicitly names the patent of Ley and Schmidt as the origin — though only after the inventor had pointed this out to the authors in February 2024. Correction is therefore possible, but it does not happen by itself.

X. What Remains

The third part of DIN 3689, which is to describe the manufacture of the H-profiles, is in preparation. It will establish the two-spindle process as the state of the art. Whether it names its origin correctly will show whether a discipline can correct its memory.

The invention is now free; the patents have expired, anyone may use it. What is missing is the same as in 1996: someone who breaks the chicken-and-egg chain and crosses the valley. The inventor is ready for it. He has waited forty years, and by now he knows exactly why it failed last time.

To those involved he can only call after them what people say in Colombia when they cannot settle a bill themselves: Dios le pague — may God repay you.