 Why would studying the patent disputes of the industrial revolution in England help us to understand the patent disputes of the digital revolution? Several hundred years separate the era of the industrial revolution from that of the digital revolution. The technology of the two eras is fundamentally different. What have the two got in common apart from they were both significant periods of innovation? Well, Oliver Wendell Holmes said the history of what the law has been is necessary to the knowledge of what the law is. And if we look back at the history of the patentability of process patents during the industrial revolution, I would argue that it may help us to understand how the patentability of computer programmes will progress. It was in England that the industrial revolution would first gain momentum. The 18th century ushered in a period of unprecedented innovation. Mechanisation inventions would encourage large-scale factory production. I have a few pretty pictures. Yes, we have innovations in the manufacture of textile. Here's an early one, James Hargreaves with his spinning Jenny. A little later, Richard Arkwright looking rather well fed and satisfied with his water frame driven by water power for spinning cotton. And then, of course, very important for everybody, Matthew Bolton and James Watt. Bolton on the left, the rather flamboyant businessman, and the extremely dower-looking Scotsman, James Watt on the right. And his beautiful machine steam engine designed by Bolton and Watt. This is also an era in which chemical discoveries were very important, not just for the textile industry, but, for example, the porteries. And this is Josiah Wedgwood's Portland vase, and Josiah Wedgwood was a very good chemist who made use of the chemical discoveries at the time to perfect his enamel colours for his portery. Before the 18th century, the granting by the crown of a patent for a new invention had been a means to encourage whole new industries into the kingdom. And the consideration for the patent had been that the patentee would introduce the manufacture and get it up and running. But already in the early 18th century, the subject matter of patents was starting to change. Rather than being whole new industries, there were starting to be new techniques. Inventions had become more technical, and the number of patents registered would also steadily increase. By the early years of the 19th century, Michelin Dean have calculated that there are about 1,000 patents in force in England, whereas in 1750 there are only about 102. If we move forward to the digital age, our so-called information age is often characterised as a post-industrial age. The factories of the 19th century, which characterised the industrial revolution, have largely disappeared from Europe and the USA. Our technology has gone digital. We live in what I often find a confusing world of tablets and internet connections and smartphones. These two young chap seems to be fairly at home in this world of digital technology. More so than I am probably. But some things do not change. Determining what kind of new invention can be patented, can be as problematic for judges today as it was back at the time of the industrial revolution. Both during the industrial revolution and the digital revolution, one form of invention has been particularly problematic for judges. Processors. When is a process patentable? Then as now it is the judges who must decide the issue by interpreting the law. In the 18th century those processes were usually mechanical or chemical and the question before the judges was whether a process as separate from a product could be patented at all. In the 20th century a process can in principle be patented but not all processes are patent eligible and which ones are can vary from jurisdiction to jurisdiction. But there is one type of process that has caused much legal debate in recent times and that's the computer programme. Computer programme is a set of commands. It is a process that directs a computer to perform certain tasks. When is a computer programme patentable? Both in the industrial revolution and the digital revolution, judges were confronted with new technology. During the industrial revolution the legal profession were increasingly asked to deal with patent specifications which had as their subject matter really quite complicated technical and chemical processes and inventors frequently complains that the judges, although they were deciding upon the validity of a patent, actually didn't really understand the technology involved. Typical is warm witness who was called to speak in front of the select committee on patents for invention in 1829. He said that a lack of familiarity with the technology led to the natural tendency of judges to decide the case before them on a point of law. That being a subject with which they are familiar rather than on the practical parts of the specification. Similarly in the 20th century judges were confronted with a new form of technology, digital technology. What was the nature of software? Was a computer programme simply coded information? Could a computer programme be good? In the St Albans case heard in 1995, so really very close to the end of the 20th century, Sir Ian Glidewell was forced to conclude there is no English authority on this question and indeed we have been referred to none from any common law jurisdiction. In an earlier hearing of that case Justice Scott Baker had been quote of a view that software probably is goods within the act. Programs are as has been pointed out of necessity contained in some physical medium, otherwise they are useless. It is not simply abstract information like information passed by word of mouth. Entering software alters the contents of the hardware. However in the Court of Appeal a distinction was made between the programme which was seen as an intangible and the disk with the programme which was seen as the tangible and it was the disk with the programme that was seen as goods. Now not all commentators see software as an intangible. It has magnetic patterns, it has electronic impulses. With respect to the patentability of computer programmes, Philip Leith points out that in software related inventions the physicality of the invention was central to the success of the early UK applications. A physical machine had to be somewhere at the heart of the patented invention. The emphasis of physicality was in his view attempting to fit the new technology of software into an already existing mental conception and the best fitting conception was that of machine. So let's first have a look at patent law in the 18th century. The role of the judges in the late 18th and early 19th century English courts was vital in the development of patent law. As there was usually no examination of a patent claim before registration that meant the first real test of the validity of a patent was by the judges if a patentee sued for the infringement of his invention. But by the mid 18th century patent disputes in the common law courts have been few and far between. There was hardly any patent law to guide the judges. As far as legislation was concerned all they had was the old statute of monopolies from 1624 which had one section on patents for new inventions and for the rest a few old cases. This left the development of patent law very largely in the hands of the judges. Now the statute of monopolies as its name suggests was primarily concerned with monopolies. Patents for new inventions formed one of the exceptions to that general prohibition of monopolies. The first and true inventor could be awarded a patent as long as the invention was for a new manufacturer and one of the provisions was not generally inconvenient. So to be patentable an invention had to be a manufacturer. But what was the meaning of the term manufacturer in the statute of monopolies? And it was that issue which would bring a lot of diverse judicial opinion to the fore. Back in the 18th century the word manufacturer was still being used to describe a whole industry. So they would talk of the woollen manufacturer and the linen manufacturer and it is possible that this was the meaning that the drafters of the statute of monopolies had in mind when they had to use the term manufacturer back in 1623. At a time when the importation of whole new industries was seen to be very important. But given the lack of precedent on how the term manufacturer should be interpreted in the cases heard in the latter part of the 18th century the meaning given to the term would depend to a certain extent upon an individual judge's interpretational style. That style could be predominantly formalistic, a literal black letter of the law approach. Or it could be predominantly teleological, purposive approach interpreting the law within a contemporary social context. A judge using a literal approach to the word manufacturer would see manufacturer in terms of something made by the hands of man. Is etymology being manus and fatura? This required a thing to have been made. On the other hand, a judge adopting a purposive, teleological approach was more likely to recognise that many patents had been granted the methods already that these methods were a great value to industry and therefore to the country and should therefore be equally deserving of protection as was a material thing. Now whether there could be patent protection for a process as separate from a product was considered in great detail in the case of Bolton and Watton Bull in 1795. Now that was one of the cases dealing with James Watt's patent for the steam engine. The defendant in that case had been accused of infringing Watt's patent. But it had been argued for the defendant that Watt's patent was not for a mechanical device but rather for a method. Indeed, Watt had himself described his invention in the patent specification as a method of lessening the consumption of steam and consequently fuel in fire engines. But said the defendant, a method was not a manufacture and therefore could not be patented and therefore there was no infringement of Watt's patent because Watt's patent was simply not valid. In Bolton and Watt the majority of the judges expressed the opinion that a process could not be the subject matter of a patent. One of the judges, Justice Buller argued, I think it is impossible to support a patent for a method only without having carried it into effect and produced some new substance. When the thing is done or produced then it becomes the manufacture which is the proper subject of a patent. So, according to that reading, a manufacture required a new physical article to be produced. There could be no patent for a process as separate from a product. And of like mine was, for example, Justice Heath. He considered that the term manufacture covered two classes. The first class was machinery and the second class was substances produced by chemical or other processes. However, the subject matter of a patent must be something that could be bought and sold. It had to be a manufacture, something tangible, a machine or a substance but something that could be bought and sold. The grant of a method, he said, is not good. Lord Chief Justice Er, however, had a very different opinion on this matter in Bolton and Watt and Bull. He emphasised that methods could be of great value to the development of industry and trade and saw no reason to exclude them from patentability. Given that the patent system at that time was one of registration rather than examination, many patents for method had already been granted. Er saw no reason, in his words, to shake the foundation upon which these patents stand. Probably I do not overrate it when I state that two thirds, I believe I might say three fourths, of all patents granted since the statute passed are for methods of operating and of manufacturing, producing no new substances and employing no new machinery. Er was aware of patents for new methods where the sole merit and the only effect produced is the saving of time and expense which lowered the price of the article and therefore introduced it into more general use. Er said, now I think these methods may be said to be new manufacturers in one of the common acceptations of the word as we speak of manufacturing of glass or anything of that kind. So, as Chief Justice Er had considered the term manufacturer did encompass a method that judges had not, it was still not entirely clear whether a method was patentable or not. A judge using a literal approach to the word manufacturer would see manufacturer in terms of a thing made by the hands of man, so he needed a thing and that was indeed seen as the meaning of the term manufacturer by Lord Kenyon in home blower and bought another case and bought steam engine which was heard a few years later in 1799. But Er was not the only judge who thought the term manufacturer could cover a method as well as a physical thing. Lord Eldon, Chancellor of the Court of Chancery, was of the same opinion. However, more commonly judges sidestep the issue by holding a process patent to be valid if it included a material component. So any instruments that have been built to put the process into effect would be seen as the manufacture although those instruments were only relevant to carrying out the process itself. Legal uncertainty on the patentability of methods persisted into the 19th century. However, Collier, who wrote the first ever legal treaties on patent law in 1803, asserted a patent cannot be granted for a method or principle. Its object must be some substantial thing produced. Although Collier did go on to note that if the patentee described his invention as a method when in fact the patent specification showed it was a tangible thing the verbal inaccuracy would not invalidate the patent. Richard Godson writing his legal treaties on patents some 20 years later than Collier and a rather more sophisticated work than Collier's was less emphatic with respect to the status of a method. That a mere method of making a thing or a process or a manner of operating cannot be the subject of a patent is not quite so clear. Much discussion has taken place on this rule. While Godson, like Collier, was convinced that the simple use of the term method would not itself invalidate a patent if the specification was actually describing something tangible he doubted that a method in itself was sufficient for a patent. A process cannot be a manufacturer within the meaning of the statute. Because it is destitute of one of the qualities absolutely necessary to be found in a new manufacturer or subject proper for a patent, materiality, the description given by that very learned judge ARCJ is not of anything that can be made. There is nothing corporal, nothing tangible nothing that can be bought or sold. No instrument by which the supposed benefit is produced and which might as an article of trade be purchased and used by another person. The legal uncertainty concerning the validity of process patents was one of the reasons that a select committee on patents for new inventions was constituted in 1829. Benjamin Roche appeared as a witness before that select committee. Roche was a barrister well known for his scientific knowledge. He'd appeared as counsel in several patent cases and was himself a patentee. Roche argued that the main source of the problems confronting patent law was the statute of monopolies itself. The words of the statute, which are extremely well calculated for those times do not happen now at all to hit the necessities of the present period. The consequence is that the judges are constantly straining the meaning of this act to make it meet the necessity of the times. Thus it depends on the extent of laxity that a judge will venture to give as to what the law at this particular day in any particular court happens to be on patents. As Roche intimated, there had been court decisions which had already de facto sanctioned a reading of the term manufacture that was broader than a literal interpretation of that term. And here are a few of them. In the 1820s and 30s various cases established that there could be a patent for processes involving a new combination of well known materials. Examples are Hall's patent for singeing off superfluous fibres of lace, Russell's patent for welding tubes, the Rossner's patent for taking the colour out of sugar and Cornish's patent for elastic cloth. A change in a process which enabled a product to be made more efficiently or more economically had been implicitly accepted by these judges as constituting the manufacture. That a method could be the subject matter of a patent was endorsed by Lord Chief Justice Tyndall in Crane and Price 1842 and this case has been seen as finally settling the law on this point. A shift in judicial interpretation had taken place and manufacture could be a process as well as a tangible object. So in the 18th, early 19th century the legal point at issue was whether a process as separate from a product could be the subject matter of a patent. By the mid 19th century a process was not by very definition excluded from patentability. But even in modern times not all processes are patentable. Article 52, two of the European Patent Convention 1973 explicitly excluded certain inventions from patentability. Inventive processes excluded from patentability included a mathematical method, a business method and a programme for a computer as such. Now a computer programme can be seen as a process as it is a computer programme that instructs and directs the hardware. The patent legislation of European states like the United Kingdom and also the Netherlands was modelled upon the EPC and these exclusions were therefore incorporated into national patent statutes. Yet the legal debate on the patent eligibility of computer programmes would prove to be far from over. The approach to the protection of a computer programme adopted in Europe would be to consider it as a literary work and hence protected by copyright. But programmers often found copyright protection insufficient. Yes it would prevent someone else from copying the original expression of the source code and the objects code. But it was often possible for a programmer to work around the copyright protection. A programmer could implement a programme's functionality in different ways, avoiding copyright. And indeed the Court of Justice of the European Union has just ruled that only the expression of the source code and the object code are protected by copyright, not the functionality of a computer programme nor the programming language. In the case of software related inventions, a patent on the other hand would give an exclusive right to apply the idea. So it's not very surprising that cases involving software related inventions made their way into court. While it was acknowledged that a computer programme could not be patented, the prohibition in the EPC and the UK Patents Act 1977 only applied to a computer programme as such. The issue then confronting the English judges, like their colleagues in Europe, was to determine when a computer programme was only a computer programme and therefore excluded from patentability, when a computer programme was something more than just a computer programme as such and not excluded from patentability. In the courtroom discussions on the meaning of the term computer programme, we see a familiar scenario, a diversity of judicial interpretation. For example, Justice Faulkner's interpretation of a computer programme in the case of Merrill Lynch's application was later disputed by the Court of Appeal in the Genotech case. In Gail's application, the Court of Appeal reached a different conclusion that had been reached by Justice Aldous in the Patents Court hearing of that case. By the time the Court of Appeal heard Aerotel and Telfer Holdings in 2006, the Court was able to observe that different approaches had been adopted over the years. Two different approaches were the so-called technical character approach and the any hardware approach. The technical character approach asks whether the invention has a technical effect, and it makes a technical contribution to the prior art. If not, it is excluded. But a technical contribution would make a computer programme more than a computer programme per se. This was the approach adopted in Gail and Frucidsin, and the approach of the English courts was similar to that adopted by the EPO Board of Appeal in Vycom. The other approach, the any hardware approach, is whether it incorporates or is implemented by some technical means such as a computer. If the claim involves the use of or is to a piece of hardware, then the exclusion would not apply. This approach had been taken by the Board of Appeal in a trio of cases, pension benefits, Hitashi and Microsoft data transfer cases, although within this trio there were variations of the any hardware approach. In the Aerotel case, Lord Justice Jacob was critical of the state of conflict which had emerged between the Boards of Appeal old approach applied in the Vycom decision and its new approach in cases like Hitashi. Jacob considered that this change in course by the Board was based upon an incorrect interpretation of what the EPC had meant by the term computer programme. He explained that one interpretation saw a computer programme as a set of instructions which could be written down on a piece of paper, an abstract thing. Another interpretation of computer programme was that the term also covered the instructions on some form of media such as a disk or hard drive which caused the computer to execute the programme. In other words, a programme in a working form. The earlier Board of Appeal decisions and the decisions of the English Court of Appeal indicated that simply incorporating a computer programme in a piece of hardware was not in itself sufficient to prevent the exclusion. A further technical contribution was needed. The trio of decisions reached by the Board in more recent cases indicated that a programme on a physical carrier was not a computer programme as such because then the computer programme was more than an abstract set of instructions. Jacob, expressing the view of the Court of Appeal, considered that the Board's new reading of the term computer programme was wrong. The exclusion in the EPC was meant to cover not just an abstract series of instructions but also a working programme, otherwise the exclusion was without real content. The Court of Appeal, in the Airtel case, rejected that any hardware will do interpretation. The patentability status of software-related inventions had become increasingly confused. That confusion was acknowledged in the Astron Clinica case heard in the Patents Court in 2008. Justice Kitchen reviewed this history of conflicting interpretations. He pointed out that the criticism of the English Court of Appeal in the Airtel case, with respect to recent decisions by the Boards of the EPO, quote, is directed at the any hardware will do approach and the return to form over substance with a drawing of a distinction between a programme as a set of instructions and a programme on a carrier. In short, he said, the Board appears to have found that any programme on a carrier has a technical character and so escapes the prohibition in article 52 following Hitachi. But he also made a particularly interesting remark. He considered that the Board was clearly influenced by the apparent illogicality of allowing claims to a suitably programmed computer and to the method performed by the computer, so programmed, but not to the programme itself. Up to the present day, there is still no one standard interpretation of article 52 EPC. Comparing the process disputes in the Industrial Revolution and the disputes on the patentability of computer programmes, several observations can be made. Each statute brings to the fore its own interpretational difficulties for judges. English judges in the 18th century had to develop the legal concept of what constituted a manufacturer in the statute of the Nautilus 1624. English judges in the 20th century had to develop the legal concept of what was meant by a computer programme as such in article 52 EPC, which formed the basis for section 1 subsection 2 of the Patents Act 1977. The law relating to the patenting of new inventions was in its infancy in the 18th century. With little in the way a developed law to guide them, it was left to the judges to try to determine whether methods were a manufacture and hence patentable. The dominant interpretation in the 18th century that a physical object was required to constitute a manufacture would give way by the early 19th century as the call to the factor accepted a process could be patentable. So too in the early years following the implementation of the Patents Act 1977 there was no well established body of case law to help judges determine the nature of a computer programme. Judges would disagree upon when a computer programme was a computer programme as such as judges applied different approaches. In 1829, Roch had considered that the statute of monopolies, a product of the 17th century was no longer in tune with the needs of society. He considered that the legal uncertainty had arisen because judges were straining the meaning of the act to make it meet the necessities of the times. The same thing could be argued with respect to the Patents Act. What had the drafters of the EPC back in the early 1970s understood a computer programme as such to be. No definition was given in the EPC as it was left to the boards of appeal and national courts to work out the detail. But the 1970s was an age when computer hardware filled a whole floor and no one had a PC. Back when software was sold and sold but was usually included on large computers or the user simply made his own programme. Back before Bill Gates Microsoft had brought the digital revolution to ordinary members of the public. Did the courts and the boards back in the 70s and 80s really understand the new digital technology or appreciate the important role that software would play in a new digital world? Is the older interpretation of a computer programme as such from that time no longer in tune with the needs of contemporary society? If the boards of the EPO in more recent times came to the conclusion that to exclude the programme itself was illogical this illogicality has arguably resulted in the EPO's Any Hardware Will Do approach. I would argue that this Any Hardware Will Do approach has its parallel in the approach to process patents by English courts in the 18th, early 19th century. At that time the underlying question was whether it was illogical to exclude from patentability a method that was clearly of great value to commerce and industry simply because the term manufacturer was being generally interpreted as referring to a material product. In determining whether a method fell within the scope of the term manufacturer the approach that would be adopted by the English courts by the early 19th century could be termed the Any Instrument Will Do approach. During the early phase of the Industrial Revolution only a few judges were prepared openly to treat the term manufacturer as covering a method separate from a product. Nonetheless, by accepting any instrument produced that was necessary for the method to be carried out as being the manufacturer a patent for a process could be upheld. Pinning the patentability of a computer programme to some form of material object hardware is very reminiscent of the reasoning of the majority of the late 18th, early 19th century judges in these process cases. Over the course of the Industrial Revolution the teleological purposive approach would triumph over a more formalistic approach to the interpretation of the term manufacturer. As Lord Mansfield stated, as the usages of society alter the law must adapt itself to the various situations of mankind. Methods of use and value to industry would be accepted as patentable. Similarly, with respect to the interpretation of a computer programme as such a teleological purposive approach will probably triumph over a more formalistic, restrictive approach. Software has become a valuable item and sometimes that software costs more than the hardware on which it is running. Commentators observe that hardware is becoming of less value than it was to the market back in the period in which the EPC was drafted and indeed one commentator has predicted the time will come when manufacturers will give away computers so as to be able to sell software. To conclude, English judges in the 18th, early 19th century rejected an earlier interpretation of the term manufacturer in the light of a new industrial social context. It is highly likely that the judges of the 21st century will adjust the legal concept of a computer programme as such to fit the social context of the digital age. Copyright protection only extends to the original expression of the source code and object code. That means the way the programme functions, the way it works is unprotected unless it can be protected by patent. Arguably a computer programme as such may then come to mean nothing more than what is covered by copyright protection. If so, the exclusion of a computer programme as such from patentability may indeed in practice become meaningless. Thank you for your time.