Copyleft and the theory of property

The history of claims to private ownership of non-material objects has passed through three stages.1

Phase 1 (sixteenth to eighteenth centuries)

The rights of creators (authors and inventors) were invented at the end of the fifteenth century.2The relevant legal framework was consolidated during the eighteenth century.3 Such rights were seen as temporary, exclusive and transferable and entitled the creator to sell copies of original works (copyright) and to make use of new inventions, provided that publication took place (patent).

During this early period, author’s rights and patents were really side issues that only emerged on the fringes of (and in the shadow of) a general theory of property that was progressively elaborated in the course of the eighteenth century and was modelled on the notion of ownership of material goods. These rights were not always spontaneously associated with “property” in the sense of “a person’s absolute rights of ownership over a material object”.4

Phase 2 (1830-1950)

The increased force accorded to such rights caused them to be associated with property rights. In France there were three views:5

a) Assimilation
Creators had ownership under natural law because what they had created was the fruit of their labours and/or the expression of their personality, in exactly the same way as for owners of material goods. Consequently, the way that laws were applied needed to be brought into line so that the absolute and permanent character of the creator’s rights was recognized.6

b) Rights sui generis
Creators had ownership under natural law but, since their property was of a special kind (having no comparable instances, no rival claims, etc.) then their rights too were of a special kind, sui generis (temporary, not involving rights of succession, etc.).7

c) Exclusive privilege
Non-material objects (speeches, ideas, etc.) cannot be held as property: their creators are not owners in natural law. Society assigns to them temporary monopolies on their use and on remuneration for the use of their works or inventions. These are privileges and not absolute rights.8

None of these views would have been imposed if the accepted definition of property had not changed. But it so happened that, towards the end of the nineteenth century, doubts began to be raised about the notion of absolute right: legal entities had become proprietors, new kinds of object had appeared and this brought about a gradual erosion of the dogma of “absolute rights of ownership of a material object”. It went so far that, in the second half of the twentieth century, people were talking, either gleefully or with horror, about the extinction9 or disintegration of ownership.10

Phase 3 (1950-2010)

This theoretical vacuum relaunched the debate over the rights of creators. If ownership no longer meant “absolute power to dispose of the object”, why could it not therefore be simply a legal guarantee of exclusivity (that is, the exclusive right of use of the item of property)? For those who supported this view, the essence of ownership was not the absolute power to dispose of the object but the exclusion of third parties. One general theory had collapsed to be replaced by another that was better constructed. According to its supporters, the dogma of ownership had now taken on its final form and there would be no phase 4.

In this new theoretical context, creators have proprietorial (that is, exclusive) rights (whether absolute ownership, rights sui generis or privilege), no matter what their legal category (copyright, patent, etc.). It is now possible to speak of “intellectual property”; indeed, the term itself is very much in vogue.11

This success is illusory: specialists are still divided on the question of the nature of these rights and the application of the law varies considerably. It is now possible to describe temporary privileges as “property”.12 On a theoretical level this amounts to a revolution: ownership now means holding exclusively rather than controlling physically and creators’ rights have become archetypal.13 Illusory though it may be, the success of the expression “intellectual property” is of great practical importance. It is now generally agreed that it contributes considerably to the extension of these rights to new objects, that it helps to strengthen the prerogatives of their owners, brings about convergence in the ways that legal systems deal with various types of creation and provides ideological support for their legitimacy (“property” sounds a lot better than “temporary monopoly”).14 This is the situation that we are in at present.

What comes next? The end of the story, or something that takes us one stage beyond intellectual capitalism? One thing is certain: whilst mature capitalism stands squarely on a platform of extensive and solid intellectual property, a bitter battle is being fought between its supporters and those who defend the idea of “common property”, of “the public domain” of what is “free” (without charge). Is this merely resistance by old forms of cooperation that have been crushed by the expansion of informational capitalism, or is it a forerunner of “a hitherto unheard-of social transformation”,15 characterized by the emergence of a form of “informational communism”?16 Perhaps the most spectacular of these conflicts is the free software movement’s challenge to manufacturers of “proprietary” software programmes.

The free software movement and property

The free software movement was begun in the 1980s by Richard Stallman and other programmers, who set up the Free Software Foundation (FSF) to write software programmes and to issue them under a kind of licence known as “copyleft”. The GNU Manifesto (Gnu’s Not Unix!) was written and published by Stallman in 1985 (and regularly updated since 1993); it sets forth the principles and aims of the project and is still the movement’s main theoretical source of reference.17 The classic view – at least, the American one – is that copyleft is to be understood as an exclusive right to sell copies of works. It is a form of copyright that the GNU Manifesto explicitly refuses to see as an “intrinsic” right, that is to say, as a natural right of ownership over the fruits of one’s labours or the expression of one’s personality. It thus subscribes to the notion of a temporary monopoly on exploitation that is granted to authors where their works are deemed to be socially useful (Cf. phase 2, c).
In other respects Stallman does not question the legitimacy of ownership of material goods. Whilst it is true that he does not make it a point of principle, the logic of his remarks would be that everyone has a right to exclude third parties from enjoyment of the fruits of his labour, that it is unjust to remove a material object from the hands of its owner. Allocation of private property rights creates a competitive system – capitalism – which would be perfectly legitimate if it always worked properly (which it does not).

In Stallman’s work there are, then, two definitions of ownership and they involve two distinct systems of legitimation: ownership of material goods as the “intrinsic” (absolute and permanent) right of workers to the exclusive enjoyments of the fruits of their labour, and “instrumental” ownership of non-material property as an exclusive privilege granted to its creators in the form of a temporary monopoly on the exploitation of their creations, to the extent that these are socially useful. On the technical level, “intrinsic” ownership is a right that is proper to the individual, that has legal force even where no title is available, whereas “instrumental” ownership implies that legislation has previously been enacted that recognizes impersonal and general statuses (“author”, “inventor”), varying slightly from one legal system to another. Any individual may lay claim to the latter form of ownership once they are in a position to state that they have created a work or invented something. Two distinct forms of ownership then, but the principal characteristic of each of them is the right to exclude third parties. Stallman is therefore a faithful representative of the dominant view to be found in Phase 3.
In Stallman’s view, copyright and patent are, each in their own way, property rights (in the sense that they are exclusive); their assignment to authors and inventors amounts to the creation of a proprietary operating system. This should make us use the word with great care. According to Stallman, the question of property can never arise in general terms: a distinction has to be made between material and non-material creations.18 The term “property” will do perfectly well for the former but there are some very good reasons for not using it to denote “exclusive privilege”, even if you qualify it with the adjective “instrumental”. The reasons are as follows:

a) The generally accepted understanding of the term “property” is that it is a “right” and rights are often considered to be intrinsic. The term “intellectual property” implies that creators’ rights are “intrinsic” in the same way as material ownership of the fruits of labour or expression of personality. They are, however, only privileges, the legitimacy of which derives from their social utility.

b) This confusion leads one to regard as self-evident the legitimacy of creators’ rights, whereas their usefulness needs to be assessed on a case-by-case basis, depending on the types of creation and the social context.

c) The American constitution provides a guarantee of protection for material property, not for “intellectual property”. Creators’ rights figure in it as temporary and relative to the social progress that they bring about. This amounts to recognition of the rights of the public over works and inventions.

d) “Intellectual property” creates the illusion that there might exist something like a “common right” governing all non-material property whereas, in reality, works, inventions, brands, etc. present specific legal problems that have to be dealt with separately if we are to have any hope of resolving them. In practice, this illusion helps to confuse these two quite different systems, with the result that they become less likely to control the processes that they were meant to regulate.

e) The illusion of “Intellectual property” leads to an unjustified alignment of the degree of protection enjoyed by owners of non-material property with that afforded to owners of material goods; this only serves to increase the profits of “creation magnates”.

Stallman notes, with some degree of sympathy, that other expressions have been suggested which are less burdened with prejudice and confusion, such as IMP (Imposed Monopoly Privileges) or GOLEM (for Government-originated Legally Enforced Monopolies). These expressions have the merit of making clear that we are dealing with privileges that are by nature distinct from rights to material property; nevertheless they still cannot avoid the reproach that they cover very different legal provisions and thereby contribute to their undesirable convergence. He therefore suggests to his readers that they should try to do without any one expression to denote the different kinds of exclusivity that creators might be granted in respect of their creations.

In terms of legal categories, Stallman has just two targets: the identification of creators’ exclusive privileges with “intrinsic” rights and the incorrect use of the term “intellectual property”. He has nothing against the idea of exclusive appropriation of non-material creations, the idea of instrumental intellectual property, providing that it does not call itself that. So we are well and truly in phase 3 – but now with an added linguistic refinement.

Books, not software programmes

This linguistic refinement is by no means innocuous: it means that books and software programmes can be treated in different ways. Stallman supports copyright being granted to the authors of books because the exclusivity that this allows makes it possible to reward authors without unduly hampering the free circulation of books and the ideas that they contain. Admittedly, this increases the sale price of the book somewhat, but on the whole the exclusivity of copyright excludes only commercial competitors. Even then it is only a temporary form of exclusion, and since it provides a simple means of encouraging authors to create, it can be seen as only mildly detrimental in terms of social utility. In the final analysis, it is a clever means of reconciling the public’s right to read books with the authors’ right to be paid a just reward for their labours.

Copyright on software programmes is not socially useful, on the other hand. In this instance the readership no longer simply consists of readers who are happy to read for pleasure, but of programmers who are themselves involved in writing new programmes by modifying (either by extending or including) the existing programmes. Software is not just a product, it is a writing process. Moreover, distribution of copies of books is an industrial activity, whereas giving a software programme to a friend is the quickest way to distribute it. Most importantly, whereas access to a book provides access to its content (the ideas that it contains), you cannot gain direct access to the contents of a software programme (the source code) because it is very difficult to decompile the source code from the object code. Finally, whereas reading may be a solitary activity, programmers work as a community. The effect of copyright is both to prevent sharing within this community for the purpose of improving software, and to prevent any collective enjoyment and efficiency arising from cooperation. Copyright on software programmes is therefore detrimental from the point of view of economic efficiency, of the freedom of the individual, and of the ethics of cooperation, selflessness and public spiritedness.

The free software movement presents itself as being “anti-proprietorial”, but in fact, Stallman

a) agrees that ownership of material goods is a natural right and he does not question its legitimacy;

b) eschews the term “intellectual property” but not an “instrumental” view of exclusive appropriation in the shape of a temporary privilege and always providing that it is socially useful;

c) considers that this proviso is met in terms of inventions and books, provides a (very rapid) discussion of brands, and categorically refuses the application of copyright to software programmes. As for the system of competition that arises from the process of granting rights to private property, namely capitalism, Stallman considers that it would be perfectly legitimate if it always worked properly. It does not, however, especially since ownership enables monopolies that ruin the system of competition. We are indeed still in phase 3, with a linguistic refinement.

This linguistic refinement is not so innocuous as to rule out the possibility of “intellectual property” (with or without the word itself) being applied to a strategic area of IT capitalism, that is, the world of software programmes. This is a position that comes close to being a particularly altruistic and public-spirited version of utilitarianism, and to this might be added the view – one that is quite widespread among lawyers19 – that copyright does not work well with software programmes. At present, there is nothing on the horizon that resembles an “alternative to IT capitalism”.20 However things get a little more interesting when we consider the question of copyleft, the alternative to “proprietary software programmes”.21

What is copyleft?

In appearance, copyleft is the opposite of copyright: “Copyleft is a general method for making a program (or other work) free, and requiring all modified and extended versions of the program to be free as well.”22 In concrete terms this “general method” involves the invention of a classification that denotes a set of original legal instruments, distribution licences or standard contracts (that are difficult to define but might be thought of as being similar to membership agreements) which govern access to and use of the work that is placed in the public domain and which are conditional upon certain principles being adhered to. The best known is the “GNU GPL (General Public Licence)” which has been the model for many others (GNU LGPL, LGL, etc.). Thus copyleft is a “general concept” that denotes a set of principles governing certain distribution licences.

The principles involve abiding by four freedoms (“freedom to run, study, distribute and improve the source code”) and one prohibition. “Run” indicates the freedom to execute the programme for any kind of use. “Study” allows free access to the source code. “Distribute” guarantees the freedom to give or to sell copies. “Improve” grants freedom to modify the programme, which again implies access to the source code. The prohibition bans any attempt to prevent modifications being left free. This fifth principle is probably the most important and the most paradoxical, inasmuch as it excludes the possibility of exclusion. Under a copyleft licence you remain free to make use of, to study, to distribute, and to modify a software programme but not to exclude third parties from any modifications on code for which you are responsible. The ban may therefore be seen as a legal requirement to grant the four freedoms to everyone. Copyleft is copyright turned upside down.

Since the movement was born, other types have appeared, the best known being “creative commons” and “open source”; both display the characteristic features of free software. There are four kinds of creative commons licence:

a) Attribution – free use but the work must be attributed. These are “free licences” (which abide by the four liberties and the prohibition).

b) Non-commercial – free use but not for commercial purposes. These licences are not “free”; they do not allow free use for commercial purposes. So copyleft is free but not necessarily free of charge, as Stallman pointed out in an image that has become famous: “free as in free speech, not as in free beer”.23 Or, to put it another way, software programmes are still merchandise like any other once they are distributed as widely as possible.24

c) No derivatives – free use but no modifications. These licences are not “free” because they do not allow the freedom to modify the software. It runs counter to copyleft which is an agglomerative provision (in that any modification can itself be modified) allowing the setting up of communities that are dynamic, even proliferating, and to which one can only belong by becoming an active participant in their proliferation.

d) Share alike – free use, including modification, conditional upon agreement to the four freedoms. These licences are “free” (they abide by the four freedoms and the prohibition).

As for Open Source, this is a dissident group within the free software movement but is only minimally different in technical terms (almost all the software programmes are free). It differs inasmuch as it rejects the political and moral dimensions of free software,25 in the sense that many projects may formally abide by some or all of the principles of copyleft but be in stark contradiction with the promotion of freedom as understood by the free software movement. For example, whilst it is true that Google makes intensive use of open source programmes, the “freedom community” sees it as a destroyer of freedom because it is capable of sacrificing respect for privacy to the interests of profit.

Thus copyleft puts itself forward as a “general method” (indeed, sufficiently general to serve as an organizational model for all areas of non-material creation26) closely associated with a radical scheme to safeguard individual freedoms and to promote communities that are productive, ludic and self-regulated. This scheme is explicitly opposed to commercial selfishness and to the sterile generation of competition that results from the adoption of proprietary systems. Just as explicitly, it refuses to take a stance against commercialization or against the private appropriation of intellectual creation in general. It is hard to imagine how the special case that was left open in phase 3 could generate an alternative to the commercialization of information and, a fortiori, to capitalist organization of non-material production. This is all the more the case given that the legal effectiveness of copyleft is based on the protection of copyright.

In the first version of the GNU Manifesto (1985), the term “copyleft” does not appear; instead, Stallman puts forward the GNU licence as a means of giving away a software programme for free use by all third parties. In terms of legal technicalities, a free licence would be a renunciation of all proprietary prerogatives. In later versions, Stallman corrects this idea: authors of free software do not renounce all their prerogatives as proprietors (except for the name), since they impose on future users quite strict clauses affecting distribution (the four freedoms and the prohibition).27 It is not so much a renunciation as a “free publication” – “release as free software”.

On the technical level the difference is considerable. Renunciation would be tantamount to putting the software in the public domain, free of all author’s rights. Such a solution would admittedly fulfil the requirements of the four fundamental freedoms but not the requirement forbidding private appropriation of subsequent modifications. The result would be, in Stallman’s view, that an “intermediary” would be free to deprive the public of access to the enhanced versions of the software by making his own modifications subject to copyright. This process comes into the category of “ownerless objects” (res nullius), which are not owned but not “unownable” and therefore open to private appropriation. To renounce the software without setting up some mechanism to prevent exclusivity would amount to transforming it into an additional resource for the “proprietary system” (alias the large companies that dominate the market, or even Microsoft). Allowing freedom is not enough: you still have to protect it from its enemies.28

In order to do so, it is no use trying to turn the weapons of the system against it. The legal structure of this “release as free” system is twofold: it is a classic copyright, to which distribution clauses are added, and takes the form of a licence to which third parties subscribe in order to have the right to use the software. The legal effectiveness of copyleft rests on an entirely classic property right (except for its name, of course); this entitles the owner to exercise the right of exclusion.29 We are very far from any idea of renunciation.

It is thus impossible to view copyleft merely as a political alternative to intellectual property or as “one step beyond it”, because it derives its regulatory force from a property right (except for the name) understood as an exclusive privilege over creations that is granted to their creators. Without copyright there can be no copyleft. It is because authors are the owners of their own creations that they have the freedom to use their property freely; this includes the freedom to decide how it shall be distributed. It must nevertheless be acknowledged that it is a strange use of the term “property”.

The “underside” of copyleft: rights, privileges, belonging

Copyleft uses ownership, in the sense of exclusivity, as a weapon to exclude exclusion. It might lead one to think of it as a form of subversion, since it involves making copyright produce the opposite effects to those that one would normally expect it to produce. However, the image (subvertere: to turn upside down) could turn out to be unfortunate: in the present case, it is the “top” (copyleft) that does the turning and not the “underside” (ownership).

“Beneath” copyleft, the first thing that we find is the freedom to choose the way in which an original piece of software is distributed. Any author of software may choose to cede his rights to the public (except for his moral rights where there is an “author’s rights” system); authors may also choose to keep for themselves in private ownership the use of what they have created (copyright), or to attach a free licence to their creation. This option is provided for them by copyright; it is because their creation is their property (with or without the name) that they are free to decide on the conditions for distributing it. As for the question as to whence copyright derives, opinions differ. Some see it as a right that is intrinsic to ownership, a right constituted by the act of creation itself, by virtue of the right to the fruits of one’s labour or the expression of one’s personality. Others, such as Stallman, see it as an exclusive privilege that takes the form of a temporary monopoly on exploitation, dependent on the social usefulness of the creation.

In the second of these views, copyright is not constituted by the creative act but granted by society. However, the creative act is not without legal force: it triggers the application of the status of author or inventor recognized in law. Indeed, all that is required is the recognition (whether implied or made explicit through a process of declaration or registration) of a creation as one’s own in order to benefit from the legal status that the legislator confers on this type of creation. Copyright and patent are statuses that support the idea of the creation as belonging to the creator as his or her own property.

Thus, the assignment of exclusive rights (copyright, patent) is based on recognition of a fundamental relationship of ownership between what is created and the creator, whether such relationship be considered as “intrinsically” legal or whether it only becomes legal through the legal status that is conferred upon it. This fundamental relation of ownership between the creator and the creation is the basis of both copyright and copyleft.

What is an author? Copyright and ownership

The relationship of ownership between creators and what is created involves two elements that are closely intertwined and yet distinct: the act of creation (writing a manuscript, painting a picture, describing a new technical procedure, etc.) and the act by which the creator recognizes that the creation is his or her own. The act of creation involves the creator’s ability to invent (ingenium), their ability to produce an original (or new) object. The act of recognition, on the other hand, involves their personal identity by relating words and actions to the existence of a self, in terms of having rather than being (not: “I am these words and actions”, but “these words and actions are mine”). The figure of the creator therefore binds inextricably together creative freedom, personal identity and ownership.

It is in the writings of Hobbes that this figure is most clearly delineated. Hobbes tells us that a “natural” person is “he whose words and actions […] are considered as his own”, whereas “a feigned or artificial person” is “considered as representing the words and actions of another”. This is why the former are known as “authors” and the latter as “actors”. He who “owneth his words and actions” is the author, even if these words are spoken by an actor, because he who “[when] speaking of goods and possessions is called an owner, in Latin dominus [or] in Greek kurios; [when] speaking of actions, is called author.”30 A person becomes “natural” or acquires a personal identity by becoming the author of the words and actions that the person recognizes as their own, that is by establishing themselves, by their own authority,31 as owners of what they do and say – or indeed, as owners of themselves, as free agents, that act and speak. Because being an author, being free and being an owner of oneself are equated with each other, the fruits of the free actions of the subject thus established belong to him or her as their personal property. It explains why such ownership extends (intrinsically or subject to the conferral of a status provided for by law) to a right of ownership in the form of a copyright or a patent.

This equation of personal identity, creative freedom and belonging to oneself may be understood in very different ways, depending on whether one regards the essence of it as residing in self-awareness (Locke), free will (Hegel), the imagination’s creation of an ideal of self (Bentham), and so on. In terms of rights, all that matters is that this equation should be assumed, so that the creative works produced and recognized may be the property of their creators.

The figure of the creator who has ownership of themselves is often associated with the origins of authors’ rights and patents somewhere between the end of the seventeenth and the beginning of the nineteenth century (phase 2). This is an illusion that arises from the confusion between the intellectual insights of a few great minds (Hobbes, Locke, et al.) and the recurrent figures that populate everyday legal writing. To judge by the latter, the “creator-owner of himself” appears to a significant extent only at the end of the eighteenth century, becoming the dominant view at the end of the nineteenth, thereby coinciding with the triumph of “intellectual property”. This means that it should be understood as the allegorical hero of phase 3, disguised as a character in phase 2 (the latter being frequently disguised as a noble savage because, as usual, heroes like to appear on the ideological stage clothed in the glories of a bygone age and, if possible, to enjoy the privilege of having the same origins). If the difference is taken into account, it becomes possible to reconstitute the genealogy of views of the creator as follows:

The figure of the artist was born in the Renaissance at the same time as the birth of authors’ rights and patents. His development coincides with “phase 1”, during which these rights have a marginal existence, overshadowed by the idea of ownership of material things. This figure, brilliantly depicted by Panofsky and Kantorowicz,32 takes its characteristics from God and then from God’s various earthly stand-ins (the Pope, kings, and so on), and imports ideas that were current in jurisprudence into the area of the liberal arts. Commentators considered that there was nothing new or original about what jurists did, because the law imitated nature, even nature’s fictions, and that only the sovereign had the extraordinary power to create new or original rules that were not a copy of anything else. From the sixteenth century onwards (but already in Dante and Petrarch), the artist, the archetype of the author, is represented as one who creates without imitating, ex ingenio, through divine inspiration, “that which was never conceived of in another’s mind”.33 This creative power, existing through no authority but its own, is borrowed via a process of “equiparation” from the creative power of God, and retains some of the attributes of that power, amongst which is the ability to confer on the creator supreme power over that which he creates.

But, in order to move from the author as sovereign to the author with ownership of himself, it remains necessary that creative freedom (ingenium), personal identity and ownership of self are woven together until they become equivalent. This was not to be conceived of or enunciated until the seventeenth century (Hobbes), or regarded as self-evident in general legal writing until the twentieth century. The artist who is sovereign and has ownership of himself may be found haunting the fringes of phase 2 (for example the idea of the Bohemian artist who goes unrecognized by the philistine bourgeois public), and becomes dominant during phase 3 by becoming an allegory of proprietors in the general sense (the long-haired entrepreneur of Silicon Valley, the new Michelangelo of the business world). He now provides the dominant identifying model that may be used by all those working under the regime of informational capitalism.34

But let us return to the question of copyleft. Thanks to Hobbes and those who followed him, we know what we mean by “creators” who own intellectual property. Can we say that Stallman’s programmers are “creators” in the sense of the foregoing definition?

The author as hacker

It is obvious that a programmer is an author, for otherwise he could not claim the copyright on which the free distribution licence he attaches to his creation is based. But he does not quite match the figure of the sovereign artist who has ownership of himself. What Stallman has written transforms the author into a hacker. A hacker is not necessarily a programmer, even if he programmes software; the notion of hacking goes very far beyond the world of IT. Indeed, the examples of it that Stallman uses derive from everyday life, from literature or music, from technology or even from schoolboy jokes.35 Nor is a hacker necessarily a pirate who breaches the security of computer systems (a “cracker”), even though it happens that some hackers are also crackers. What gives value to a hack is not transgression but a subtle mixture of technical skill, dexterity in play, elegance and wit. A hacker is certainly not legalistic, but he is not compulsively transgressive either. He is content simply to extend his artistic independence as far as indifference to the rules in force.

The definition of hacking relates to ideas about play, intelligence and exploration of the limits of the possible. A hacker likes to resolve a problem that appears insoluble in the most playful and ingenious manner possible. But the aim of hacking is above all to produce high “hack value” and to be recognized within one’s community as having done so. This is why the hacker’s pleasure has to be shared with other people who appreciate the art of doing difficult things with elegance, detachment and cheek. A hacker is not an isolated creator who delivers his works to an abstract, universal public; he is a member of a group and without that group his hacking would be pointless and worthless.

Free action is here not thought of simply as creation, but as a creative game in which the creative aspect is not so much to bring into existence an original and new product; it is rather the way in which one plays it. Indeed, a hack may quite well produce nothing at all providing that the act of creating that nothing is itself creative. The accent has shifted then: it is no longer to do with the demiurgic power, which is of divine (or papal or royal) origin, but with the imaginativeness of an activity, whether productive or otherwise. The value of the hack is therefore not measured in terms of the quantum of usefulness (no matter what definition of usefulness one adopts), or in terms of the average time required for producing the object that is created, but in the degree of skill and comedy of the creative act itself.

It is not possible to objectivize this value, since we are in the realm of play, an area that is incapable of being pinned down ontologically. Thanks to Freud, we know that play is the opposite of reality,36 and thanks to Winnicott, that it occupies an intermediate or transitional zone between the subjectivity and the objectivity of the word in (and with) which he plays.37 The only universal equivalent that could be capable of prevailing in that sphere is the enjoyment of the game, and the only rules imposed are those that the players themselves invent in order to enjoy the game. A hack can therefore not be exchanged for some monetary value (to try to buy a hack would destroy its hack value). It is the reputation that the hacker gains within his or her community that represents the reward for the hacker’s talents.

In this sense, the activity of a hacker is essentially collective and makes sense only on the playing field made up of the community of those who gain mutual pleasure from admiring their hacks. This, of course, is the reason why the hacker’s “work” is less an object or a product expressing the creative freedom of its author than the setting up of an open process of collective creation that creates its meaning as a “work'” from within itself.38 A hacker is indeed an author, but the version of authorship he or she puts forward consists in incarnating a form of creative freedom that is playful, collective and disinterested. The more he or she plays at not being a sovereign proprietor, the more that is what he or she is: the carnival representation of an author. It may be said that we are simply dealing with a pose, with a performance or with a game and that none of it is really serious. It is true that copyleft does not subvert copyright but is content to play at subverting it, without balking at making use of its legal effects. But this pretended subversion makes it possible to see an entirely serious conceptual dissociation between ownership and exclusivity.

Ownership against exclusivity, the dawn of phase 4?

Whether copyright is seen as an intrinsic right or a privilege that is granted, it presupposes that its source relationship is one of belonging (creation means belonging) that confers on the creator an exclusive right over his or her works (belonging means exclusive right). Stallman faithfully goes along with the first equation (the hack belongs to the hacker) but not the second, since if the creator has the power to decide upon the conditions for publication of his or her creations, he or she is free to make use of this power to include rather than exclude third parties.

What copyleft shows is that having your creations as your private property confers the power to determine the conditions of publication, but that this power should not be confused with the power to exclude, or even that the progression from belonging to property is not necessarily a straight line.

The creator has the inherent power to destroy his or her works, to keep them secret, to abandon them, to keep the use of them to him or herself alone, to share them within a community that contains the entire range of intermediate options thanks to the great diversity in available licences. He or she has this power because it is inherent in his or her freedom to exclude, which is guaranteed to him or her by law, but which he or she is under no obligation to use. Ownership and exclusivity are inherently well and truly confused.

But, as soon as this original creator makes the choice not to exclude, by sharing his or her work within a community, for example, this choice is irreversible. Unlike the owner of a garden who can open it to the public for a day and then close it up in the evening, the owner of a software programme, because the information is disseminated, can no longer control it or prevent use subsequent to the “free publication” of his or her work. He or she remains the owner of the software in a full sense, but has little more than a formal right to exclude, one which is practically unusable.39

Choosing inclusion is tantamount to a definitive, de facto renunciation of the right to exclude. In other words, the original creator, without ceasing to be the proprietor, has the right to renounce, once and for all, the exercise of his or her right to exclude, or even: he or she only has the right to exclude so long as he or she continually renews his or her initial choice to keep for himself certain uses of his or her creation. The right to exclude is therefore not so much a “substantial attribute” of ownership as an option offered to the original owner as long as it is constantly confirmed. Ironically and by contrast, it is the “right to include” that appears more “substantial” since the original owner always has the option to choose sharing even if he might initially have opted for an exclusive method of distribution.

As for the creator of subsequent modifications to an original software programme that is subject to a free licence, his or her situation is even more radical. He or she is the owner of his creations but has never had the right to exclude, because he or she renounced it by gaining access to the source code of the original software before his or her ownership became established. In this case ownership and exclusivity are strictly and rigorously dissociated.

Copyleft faithfully goes along with the anthropological premises (the creator’s ownership of self) and the legal provisions (intellectual property, but without using that name) of IT capitalism, but it offers a rather playful version of these premises in order to bring to light the absence of connection between recognition of anthropological essence and the right of ownership, understood as exclusivity. Once having something as your personal property does not mean that you are the sole proprietor (in the sense that you can exclude third parties) – contrary to what legal tradition normally preaches, individual and common are not necessarily contradictory – then there is nothing to prevent one conceiving of inclusive ownership.

Copyleft is thus not an all-inclusive alternative to ownership; it is an ingenious arrangement that prevents any exclusivist view of ownership being elevated to the status of legal dogma. Copyleft makes it possible to visualize coexistence and expression of proprietary systems and communities freed from proprietary systems, a little like the way in which free towns in the Middle Ages were able to free themselves from certain forms of taxation,40 or the way in which certain philanthropists made use of their fortunes to found communities and phalansteries. So it is less a subversion of ownership than a neutralization of those effects of exclusivism judged to be politically and morally damaging.

Of course, one might find all of this a little disappointing. One might have hoped, for example, that the philosophical aim would be to get away from the perspective dictated by the equation of personal identity with creative freedom and ownership of self, and instead conceive of the creative act freed from the toils of ownership and identity. To achieve this it is necessary to acknowledge that copyleft is not much help: a hacker is a proprietor who is inclusive, playful and joyful – but he is still a proprietor. Should that be a matter for regret?

That is certainly the case for supporters of “anti-copyright” or “creative anti-commons”, who believe that the internal contradictions of copyleft can be resolved by getting rid of rights completely.41 It is true that copyleft is at the same time inclusive and exclusive when it forbids private appropriation of material based on a free software programme. Excluding exclusivity is still excluding. But basing an argument on this paradox, in order to demand that all rights be abolished in order to reject all forms of exclusivity, amounts to forgetting that creations that are abandoned are in abeyance, open to all kinds of re-use and to private appropriation of anything derived from them. It is not easy to understand how presenting free resources to proprietary systems is likely to hasten the end of informational capitalism.

To conclude, copyleft does not sound the death-knell of intellectual property, and probably does not even approach such a thing. It does, however, indicate the limitations of any theory that would come too close to confusing it with exclusivism; and it does make clear the inclusive use that can be made of it. Copyleft suggests that, with a little legislative imagination, it can be of use in a wide variety of situations, explicitly breaking away from the competitive dynamics of IT capitalism. It is not, perhaps, the end of a world, but it does provide food for thought.