Stevens v Kabushiki Kaisha Sony Computer Entertainment and Others

(2005) 221 ALR 448
[2005] HCA 58

---

(Judgment by: Gleeson CJ, Gummow J, Hayne JJ, Heydon JJ)

Stevens v
Kabushiki Kaisha Sony Computer Entertainment and Others

Court:

Judges: Gleeson CJ
McHugh JJ
Gummow J
Kirby JJ

Hayne JJ
Heydon JJ

Judgment date: 6 October 2005

Canberra

---

Judgment by:
Gleeson CJ
Gummow J
Hayne JJ
Heydon JJ

[1] With effect from 4 March 2001, the Copyright Amendment (Digital Agenda) Act 2000 (Cth) (the Amendment Act) made significant amendments to the Copyright Act 1968 (Cth) (the Act). This appeal concerns a dispute as to the construction of the "circumvention device" provisions introduced by the Amendment Act.

The scope of copyright law

[2] Over a long period amendments to copyright law have comprised legislative solutions to problems created by competing economic and social pressures associated with the development of new technologies. The issues in the present appeal indicate that this is very much the case today.

[3] The well-established categories of original works of authorship have been supplemented by various types of "subject-matter other than works" (including, significantly for this case, cinematograph films), certain "moral rights" have been conferred on individuals, and computer programs have been protected as literary works. This last step has been taken notwithstanding any incongruity in treating computer programs as literary works given "their objective of making hardware function rather than of conveying anything immediately perceivable to humans". ^[1]

[4] Copyright in both works and other subject-matter remains defined in the Act primarily in terms of the doing (or the authorising of the doing) of any of various acts listed as those comprised in the relevant copyright. ^[2] Other infringement provisions include those dealing with importation for sale and hire (ss 37, 102) and sale and other dealings (ss 38, 103). It follows from this specificity that not all activities involving the use of copyright material require a licence to escape infringement. ^[3] (Patent law has operated more broadly, with the traditional terms of the grant of monopoly being to "make, use, exercise and vend" the invention; however, no patent rights were claimed in this litigation.) Merely to read a copy of a book is not to infringe the literary work of which the book is a material reproduction. Further, the making available of means of reproduction which may or may not amount to infringement has been held not necessarily to amount to authorisation of infringement. ^[4]

[5] This litigation turns upon the construction of provisions in the Amendment Act which expand neither the existing categories of copyright works and other subject-matter protected by the Act nor the categories of infringement. Rather, the legislation in question deals with "anti-spoiler devices" which would allow the side-stepping of technical barriers to copying.

Anti-spoiler devices

[6] There is considerable controversy in Australia and elsewhere concerning the proper scope of such legislation. ^[5] However, the task of the court on this appeal is to construe the particular compromises reflected in the terms of the Amendment Act.

[7] The development of technical barriers to copying and the escalation of a struggle between those who design such barriers and those who devise means of surmounting them is not new. Professor Cornish writes: ^[6]

Back in the 1970s and 1980s, the answer to analogue copying on photocopiers, cassette decks, and video recorders was pronounced to lie in the machines themselves: but the eternally springing hopes were often enough dashed. Every locked door seemed to produce a hacker with a jemmy. With the Internet, technical control remains the core objective, because it seems the only hope for preserving the copyright industries in something resembling their present form. [Original emphasis]

[8] A legislative response to problems identified in the pre-internet age had been made in the United Kingdom in s 296 of the Copyright, Designs and Patents Act 1988 (UK) (the 1988 UK Act). ^[7] Section 296(2) used the broadly stated expression "any device or means specifically designed or adapted to circumvent the form of copy-protection employed". Section 296(4) defined the phrase "copy-protection" as including "any device or means intended to prevent or restrict copying of a work or to impair the quality of copies made". The distinction between preventing or restricting copying of a work and the impairment of the quality of copies made remains important in considering the construction of the current Australian legislation.

The Amendment Act

[9] Section 3 of the Amendment Act sets out what are stated to be the objects of that statute. These objects are expressed largely by reference to the internet and online access to copyright material. According to the Revised Explanatory Memorandum to the Senate on the Bill for the Amendment Act (the explanatory memorandum), ^[8] "the keystone" to the reforms provided by the Bill was the introduction ^[9] of a "new technology-neutral right to communicate literary, dramatic and musical works to the public". This would "provide copyright owners with greater protection for their material in the new digital environment". However, the substantive provisions of the Amendment Act with which this appeal is concerned deal with a different matter, technical control of "access".

[10] The Amendment Act inserted Div 2A (ss 116A-116D) in Pt V of the Act. Part V is headed "Remedies and offences". Division 2A is headed "Actions in relation to circumvention devices and electronic rights management information". The Amendment Act also introduced additions to the offence provisions contained in Div 5 (ss 132-133A) so as to create new offences for contravention of the new Div 2A. The Amendment Act further introduced new definitions into s 10 of the Act. The Act has been further amended on five occasions, the last set of changes being those made with effect from 1 January 2005 by the US Free Trade Agreement Implementation Act 2004 (Cth) (the 2004 Act). This litigation is concerned with the statute in what is to be taken as its form at the date of commencement of the Amendment Act, 4 March 2001.

[11] In the explanatory memorandum ^[10] it was said that the provisions of Div 2A were intended to provide "appropriate measures for the enforcement of copyright in the digital environment" and to provide "effective civil remedies against the abuse of technological copyright protection measures". In particular, Div 2A provided copyright owners "with new civil remedies against persons who make, commercially deal in, import, advertise, market or make available online devices, or provide services, used to circumvent technological copyright protection measures".

[12] The explanatory memorandum ^[11] also stated that the changes made were intended to ensure that Australia provided adequate legal protection and effective legal remedies to comply with "the technological measures obligations" in two treaties negotiated in 1996 in the World Intellectual Property Organization (WIPO). One of these was the WIPO Copyright Treaty, which became effective on 6 March 2002.

[13] Article 11 of the WIPO Copyright Treaty stated:

Contracting Parties shall provide adequate legal protection and effective legal remedies against the circumvention of effective technological measures that are used by authors in connection with the exercise of their rights under this Treaty or the Berne Convention and that restrict acts, in respect of their works, which are not authorized by the authors concerned or permitted by law.

It will be apparent that the provision is expressed in broad terms, leaving considerable scope to individual states in deciding on the manner of implementation. ^[12]

[14] The explanatory memorandum further said of the new Div 2A: ^[13]

These provisions will operate to provide copyright owners and their licensees with an effective means of enforcing their rights in the online environment whilst simultaneously allowing for the operation of some exceptions to the exclusive rights of copyright owners. In this way, the provisions are intended to strike a fair balance between the rights of copyright owners and the rights of copyright users.

[15] The contrast between legislation such as Div 2A and the protection of copyright in works and other subject-matter was drawn by the Committee on Commerce of the House of Representatives of the United States Congress when considering legislation proposed to amend the copyright law and to implement in the United States Art 11 of the WIPO Copyright Treaty. The United States legislation that resulted, the Digital Millennium Copyright Act 1998 (US), amended Title 17 of the United States Code, introducing inter alia 17 USC §1201, to which more detailed reference will be made below. With respect to the previous copyright provisions, that committee said in Pt 2 of its report: ^[14]

In general, all of these provisions are technology neutral. They do not regulate commerce in information technology, ie, products and devices for transmitting, storing, and using information. Instead, they prohibit certain actions and create exceptions to permit certain conduct deemed to be in the greater public interest, all in a way that balances the interests of copyright owners and users of copyrighted works.

[16] The committee went on to refer to the use of the term "para copyright" to identify anti-circumvention provisions, liability under which could result from conduct independent of any act of infringement or of any intent to promote infringement. ^[15]

[17] Before turning to consider the submissions respecting the construction of the definition of "technological protection measure" in the Act, it should be observed that the broad terms of Art 11 of the WIPO Copyright Treaty have supported legislation of various countries which is in differing forms. For example, in the United States, 17 USC §1201, which is headed "Circumvention of copyright protection systems", deals with the matter in different terms from those of Div 2A in the Australian legislation. Section 1201(a)(1)(A) states that "[n]o person shall circumvent a technological measure that effectively controls access to a work protected under this title". Then, §1201(a)(3) provides:

As used in this subsection:

(A)

to "circumvent a technological measure" means to descramble a scrambled work, to decrypt an encrypted work, or otherwise to avoid, bypass, remove, deactivate, or impair a technological measure, without the authority of the copyright owner; and

(B)

a technological measure "effectively controls access to a work" if the measure, in the ordinary course of its operation, requires the application of information, or a process or a treatment, with the authority of the copyright owner, to gain access to the work.

[18] The Australian legislative materials identified in the written submissions to the court indicate that proposals were made by the International Intellectual Property Alliance to the House of Representatives Standing Committee on Legal and Constitutional Affairs in favour of the adoption of legislation in terms such as those of the United States, with emphasis upon protection for all devices controlling access to a work. However, as will be apparent, the legislation as enacted takes a different form, with an emphasis upon measures designed to prevent or inhibit infringement.

The facts

[19] The present respondents (collectively described as "Sony") produced and sold computer games on CD-ROMs for use with PlayStation consoles. Sony as owner or exclusive licensee controls the copyright in the computer programs (as literary works under the Act) and in the cinematograph films (as subject-matter other than works) embodied in the CD-ROMs for the games.

[20] On two occasions after the commencement of the Amendment Act, the appellant, Mr Stevens, sold unauthorised copies of PlayStation games. The games were titled "Croc 2", "Medi Evil", "Motor Races World Tour" and "Porsche 2000". Mr Stevens was not sued for any acts on his part that might have constituted infringements of Sony copyright in any computer program or cinematograph film. Nor were the makers of the unauthorised copies, whether Mr Stevens or others.

[21] However, the PlayStation software contained access restrictions described as follows by Sackville J in his judgment at first instance: ^[16]

[46] The PlayStation software incorporates an access code, or a number of encrypted sectors of data that cannot be reproduced by conventional CD recording or copying devices (usually referred to as "burning" mechanisms). The access code is stored on an encrypted portion of the CD-ROM and essentially consists of a string of characters. This string must be read by the boot ROM located within the PlayStation console if the particular game is to be played. The boot ROM recognises whether there is an access code and specifically what kind of access code it is. The access code is inaccessible to standard CD-ROM "burners" or standard CD replication manufacturing parts.

Sony contended that, in this state of affairs, a "technological protection measure" could be said to exist in the boot ROM, or the access code in the PlayStation software, or the two in their combined operation.

[22] In addition to supplying the unauthorised copies, Mr Stevens on three occasions sold and installed "mod chips" into PlayStation consoles. The unauthorised copies could not be played upon an "unchipped" or unmodified PlayStation console because they did not have the requisite access code. However, these copies could be played upon the "chipped" PlayStation consoles which Mr Stevens had modified.

[23] By proceedings instituted in the Federal Court, Sony alleged that contrary to s 116A (inserted in the Act by the Amendment Act) Mr Stevens without permission had knowingly sold or distributed a "circumvention device" which was capable of circumventing or facilitating the circumvention of a "technological protection measure" which protected Sony's copyright in literary works (computer programs) and cinematograph films.

The litigation

[24] By its application in the Federal Court, Sony sought against Mr Stevens a declaration, damages, an injunction and civil relief under the civil remedies provision in s 116D. Sackville J held that the claims by Sony under Div 2A failed. ^[17] In the Full Court, Sony succeeded on the first of three issues, but not on the second or third. ^[18] However, Sony's success was sufficient to entitle it to substantial relief against Mr Stevens. The Full Court made a declaration as follows:

On 8 April 2001, 28 September 2001 and 16 November 2001 [Mr Stevens] sold circumvention devices, as defined in [the Act, s 10(1)], for use in association with "PlayStation" computer consoles and the CD-ROMs for "PlayStation" computer games, in contravention of s 116A of [the] Act.

[25] The Full Court enjoined Mr Stevens from selling circumvention devices for use in association with those computer consoles and CD-ROMs in contravention of s 116A of the Act. It remitted the matter to the primary judge for determination of the claims for damages pursuant to s 116D of the Act.

[26] Against those orders, Mr Stevens appeals by special leave to this court. By notice of contention, Sony seeks to reagitate the issues on which it did not succeed in the Full Court. At first instance, Sackville J had permitted the Australian Competition and Consumer Commission (the ACCC) to appear as amicus curiae and to press for a construction of the relevant provisions of the Act at odds with that favoured by Sony. ^[19] An application to this court by the ACCC was withdrawn. However, the court granted leave to appear as amici curiae to the Australian Digital Alliance Ltd and the Australian Libraries Copyright Committee.

Section 116A

[27] Section 116A(1), so far as immediately material, states that the section applies if "a work or other subject-matter is protected by a technological protection measure" and a person without the permission of the owner or exclusive licensee thereof makes, sells or offers for sale or hire or otherwise promotes or advertises "a circumvention device" which is capable of circumventing, or facilitating the circumvention of, that "technological protection measure". Making and importing are also proscribed by s 116A(1), but the mere use of a circumvention device is not proscribed. Supplying, making and importing are excused if "for use" for a "permitted purpose".

[28] The terms "circumvention device" and "technological protection measure" are defined in s 10(1). ^[20] Save as to what follows, it was not disputed that Mr Stevens had sold "circumvention devices". The definition is as follows:

"circumvention device" means a device (including a computer program) having only a limited commercially significant purpose or use, or no such purpose or use, other than the circumvention, or facilitating the circumvention, of an (sic) technological protection measure.

[29] What was in issue was the existence of the "technological protection measure" identified in the concluding words of the definition of "circumvention device". It is upon the following definition of "technological protection measure" that the appeal by Mr Stevens turns. The definition states:

"technological protection measure" means a device or product, or a component incorporated into a process, that is designed, in the ordinary course of its operation, to prevent or inhibit the infringement of copyright in a work or other subject-matter by either or both of the following means:

(a)

by ensuring that access to the work or other subject matter is available solely by use of an access code or process (including decryption, unscrambling or other transformation of the work or other subject-matter) with the authority of the owner or exclusive licensee of the copyright;

(b)

through a copy control mechanism.

Statutory construction

[30] The critical task for the outcome of this appeal is one of statutory interpretation, particularly of the defined expression "technological protection measure" as it appears in the setting of Div 2A. No particular theory or "rule" of statutory interpretation, including that of "purposive" construction, can obviate the need for close attention to the text and structure of Div 2A.

[31] Lord Renton QC ^[21] writes of the position in the United Kingdom: ^[22]

I do not know to what extent judicial interpretation influences drafting but drafting greatly influences judicial interpretation of statutes. From earliest times judges have found it difficult to interpret them, and most of the time of appellate judges is now taken up in doing so. Parliament has never required the judges to do so in any particular way. The Interpretation Act [1978 (UK)] merely provides some definitions and minor assumptions. So the judges have made their own well-known rules of interpretation.

Of these rules of interpretation Lord Renton continues: ^[23]

The first was the Mischief Rule in 1584: to find out the intention of Parliament it was necessary to discover the mischief for which the common law did not provide and what was the remedy Parliament chose to cure it. That rule still applies where relevant. Later came the Golden Rule, which said that, if the whole statute leads to inconsistency, absurdity or inconvenience, the court should give it another meaning that makes more sense. This caused problems and led to the Literal Rule: if the words of the statute which apply to the case being tried are clear, they must be followed, however unjust the result. Then came "the Diplock principle", that the court must give effect to what the words would mean to those whose conduct the statute regulates.

He then indicates that in England in the past 50 years the judges have gradually adopted the "purposive rule"; under this the judges try to discover what parliament intended. In Australia, s 15AA of the Acts Interpretation Act 1901 (Cth) states:

In the interpretation of a provision of an Act, a construction that would promote the purpose or object underlying the Act (whether that purpose or object is expressly stated in the Act or not) shall be preferred to a construction that would not promote that purpose or object.

Section 15AB provides for the use of a wide range of extrinsic materials in pursuing the construction indicated in s 15AA.

[32] In the case of the Amendment Act, there is a statement of objects in s 3. However, as indicated earlier in these reasons, that statement of objects, which fixes upon the "online" environment of the internet, does not encompass the broader operation of Div 2A, as demonstrated by the facts in this litigation. Nor do the extrinsic materials give any clear indication of how it came to be that the Bill for the Amendment Act took the final form that it did. Indeed, the very range of the extrinsic materials, with shifting and contradictory positions taken by a range of interest holders in the legislative outcome, suggests that the legislative purpose was to express an inarticulate (or at least not publicly disclosed) compromise.

[33] There is force in the statement by one commentator: ^[24]

The definition of "technological protection measure" is a compromise, which was neither as restrictive as some copyright users had hoped, nor as broad as copyright owners sought - and parts of the legislative history are opaque.

[34] The result is that in the present case to fix upon one "purpose" and then bend the terms of the definition to that end risks "picking a winner" where the legislature has stayed its hand from doing so. In the selection of a sole or dominant "purpose", there is a risk of unintended consequences, particularly where, as here, the substratum of the legislation is constantly changing technologies.

Technological protection measure

[35] These considerations indicate the approach to construction evident in the reasoning of Sackville J, with its close attention to text and structure. Of the expression "technological protection measure", his Honour said: ^[25]

[113] The definition has a number of elements, as follows:

•

a device or product, or a component incorporated into a process.

•

that is designed

•

in the ordinary course of its operation

•

to prevent or inhibit the infringement of copyright in a work [or other subject-matter]

•

by either or both of two particular means.

The two particular means of preventing or inhibiting the infringement of copyright are these:

•

ensuring that access to the work is available solely by use of an access code or process with the authority of the owner or licensee; or

•

a copy control mechanism.

[36] Sackville J did not accept the construction advanced by Sony which was to be accepted in the Full Court and which is urged again on this appeal. His Honour rejected the proposition that: ^[26]

[115] ... the definition is concerned with devices or products that do not, by their operations, prevent or curtail specific acts infringing or facilitating the infringement of copyright in a work [or other subject-matter], but merely have a general deterrent or discouraging effect on those who might be contemplating infringing copyright in a class of works, for example by making unlawful copies of a CD-ROM.

Rather, Sackville J said: ^[27]

[114] It can be seen that the focus of the definition, as the expression "technological protection measure" itself implies, is on a technological device or product that is designed to bring about a specified result (preventing or inhibiting the infringement of copyright in a work) by particular means. Each of the specified means involves a technological process or mechanism. The means identified in para (a) is an access code or process that must be used to gain access to the work. The means identified in para (b) is a "copy control mechanism".

[37] That latter expression is not defined in the legislation. However, the distinction between devices or means designed to prevent any copying at all and those designed to impair the quality of copies that are made has a provenance in s 296 of the 1988 UK Act, to which reference has been made at [8] of these reasons. Consistently with this and with reference to the Australian legislative history, Sackville J concluded that the phrase "copy control mechanism" encompassed a mechanism restricting the extent (and, one might add, the effectiveness) of copying of a work that otherwise could be undertaken by someone with "access" to the copyright material. ^[28]

[38] Sackville J concluded that: ^[29]

[115] ... a "technological protection measure", as defined, must be a device or product which utilises technological means to deny a person access to a copyright work [or other subject-matter], or which limits a person's capacity to make copies of a work [or other subject-matter] to which access has been gained, and thereby "physically" prevents or inhibits the person from undertaking acts which, if carried out, would or might infringe copyright in the work [or other subject-matter].

That construction should be accepted.

[39] It is important to understand that the reference to the undertaking of acts which, if carried out, would or might infringe, is consistent with the fundamental notion that copyright comprises the exclusive right to do any one or more of "acts" primarily identified in ss 31 and 85-88 of the Act. The definition of "technological protection measure" proceeds on the footing that, but for the operation of the device or product or component incorporated into a process, there would be no technological or mechanical barrier to "access" the copyright material or to make copies of the work after "access" has been gained. The term "access" as used in the definition is not further explained in the legislation. It may be taken to identify placement of the addressee in a position where, but for the "technological protection measure", the addressee would be in a position to infringe.

[40] This construction of the definition is assisted by a consideration of the "permitted purpose" qualifications to the prohibitions imposed by s 116A(1). First, s 116A(3) provides that, in certain circumstances, the section does not apply in relation to the supply of a circumvention device "to a person for use for a permitted purpose". The term "supply" means selling the circumvention device, letting it for hire, distributing it or making it available online: s 116A(8). Secondly, s 116A(4) states that the section in certain circumstances does not apply in relation to the making or importing of a circumvention device "for use only for a permitted purpose".

[41] The expression "permitted purpose" in subss (3) and (4) has the content given it by subs (7). This states that for the purposes of s 116A, a circumvention device is taken to be used for a permitted purpose only if two criteria are met. The first criterion is that the device be "used for the purpose of doing an act comprised in the copyright in a work or other subject-matter " (emphasis added). The second criterion is that the doing of that act otherwise comprised in the copyright is rendered not an infringement by reason of the operation of one or more of the exculpatory provisions then set out. ^[30] (The listed provisions do not include the general fair-dealing exculpations in ss 40, 41 and 42 of the Act.)

[42] The first criterion in s 116A(7) for reliance upon the permitted purpose provisions which are an answer to what would otherwise be a claim under s 116A thus in terms links the use of a circumvention device to the doing of one or more of the acts enumerated in s 31 of the Act (where these are done in relation to a work) and in ss 85-88 (where these are done in relation to subject-matter other than a work).

[43] If the construction of the definition for which Sony contends were accepted despite the linkage specified in s 116A(7) between the use of a circumvention device and the central provisions of ss 31 and 85-88 of the Act, the permitted purpose provisions would risk stultification. The facts of the present case are in point. The use of Mr Stevens' mod chip in order to circumvent the protections provided by

(a)

the access code on a CD-ROM in which a PlayStation game is stored and

(b)

the boot ROM device contained within the PlayStation console

cannot be said to be for the "purpose" of reproducing a computer game within the sense of s 31 of the Act. Any such reproduction will already have been made through the ordinary process of "burning" the CD-ROM. The mod chip is utilised for a different purpose, namely to access the reproduced computer program and thereafter visually to apprehend the result of the exercise of the functions of the program.

[44] There are three other considerations which support Sackville J's construction of the definition.

[45] The first is that, in choosing between a relatively broad and a relatively narrow construction of legislation, it is desirable to take into account its penal character. The present litigation does not arise from the institution of criminal proceedings under the offence provisions now contained particularly in s 132 of the Act. However, a person who makes or sells a circumvention device (s 132(5B)) is liable to imprisonment for not more than 5 years (s 132(6A)). An appreciation of the heavy hand that may be brought down by the criminal law suggests the need for caution in accepting any loose, albeit "practical", construction of Div 2A itself.

[46] The second consideration is that the true construction of the definition of "technological protection measure" must be one which catches devices which prevent infringement. The Sony device does not prevent infringement. Nor do many of the devices falling within the definition advanced by Sony. The Sony device and devices like it prevent access only after any infringement has taken place.

[47] The third consideration is that in construing a definition which focuses on a device designed to prevent or inhibit the infringement of copyright, it is important to avoid an overbroad construction which would extend the copyright monopoly rather than match it. A defect in the construction rejected by Sackville J is that its effect is to extend the copyright monopoly by including within the definition not only technological protection measures which stop the infringement of copyright, but also devices which prevent the carrying out of conduct which does not infringe copyright and is not otherwise unlawful. One example of that conduct is playing in Australia a program lawfully acquired in the United States. It was common ground in the courts below and in argument in this court that this act would not of itself have been an infringement. ^[31]

The Full Court's reasoning

[48] However, on appeal, the Full Court accepted the construction advanced by Sony. In doing so, the Full Court did not refer to the significance, for construction of the definition of "technological protection measure", of the permitted purpose provisions. The reasoning of the judges in the Full Court varied. Lindgren J, with whom Finkelstein J agreed on this issue, ^[32] found nothing in the statutory text to persuade him strongly to one construction or the other. ^[33] His Honour was persuaded by a detailed review of the extrinsic materials that a "broader approach" was intended by the parliament so that the definition of "technological protection measure" embraced an "inhibition" which was indirect and operated before an attempted operation of the circumvention device. ^[34]

[49] However, if one thing appears from a consideration of the Australian and international materials it is that in Australia there was a reluctance to give to copyright owners a form of broad "access control". Indeed, this reluctance is manifest in the inclusion in the definition of "technological protection measure" of the concept of prevention or inhibition of infringement.

[50] This outcome dissatisfied copyright owners. Yet other "stakeholders" with their own interests did not achieve all they may have desired. To those, such as the ACCC, concerned with the operation of restrictive trade practices law, it was significant that the access code for Sony products differed in various markets, so that a PlayStation game purchased in the United States could not be played on an unmodified PlayStation console purchased in Australia. ^[35] Users of copyright material such as those represented in the amici curiae in this court were dissatisfied by the exclusion from the permitted purpose provisions of the general provisions protecting fair dealing. Other users were dissatisfied by the failure to include in the permitted purpose provisions the specific protection given by s 47C for back-up copies of computer programs. All of these considerations suggest no particular support for the "broad" approach to the definition of "technological protection measure".

[51] French J, the other member of the Full Court, emphasised that s 116A operates with respect to the range of acts which may constitute infringement, a range going beyond reproduction. His Honour said of that range of acts: ^[36]

[20] It extends to knowingly selling or offering for sale articles, the making of which constituted an infringement of copyright (s 38) ... If a device such as an access code on a CD-ROM in conjunction with a Boot ROM in the PlayStation console renders the infringing copies of computer games useless, then it would prevent infringement by rendering the sale of the copy "impracticable or impossible by anticipatory action".

[52] However, the provision in s 38 which, subject to the fair dealing and related provisions, renders it an infringement of copyright in a work to sell an article "if the person knew, or ought reasonably to have known, that the making of the article constituted an infringement" may be accommodated within the operation of s 116A without going so far as did French J in construing the definition of "technological protection measure". Taking the example discussed by French J, s 38 itself indicates that what might be called the act of secondary infringement by sale must necessarily follow in a temporal and practical sense from the primary infringement of making the article. The "technological protection measure", consistently with the construction accepted by Sackville J, prevents the act of primary infringement in an immediate sense. It also thereby "inhibits" the act of secondary infringement. One meaning of "inhibit" is to hinder, to check or to place an obstacle or impediment to a path of conduct. ^[37]

[53] French J went on to conclude ^[38] that the construction proffered by Sony flowed from a consideration of the ordinary and grammatical meaning of the language of s 116A and the definition of "technological protection measure". To accept the contrary construction would be "to cage the ordinary meaning of the words which have been adopted". ^[39]

[54] Copyright legislation, both in Australia and elsewhere, gives rise to difficult questions of construction. Given the complexity of the characteristics of this form of intangible property, that, perhaps, is inevitable. It may be going too far to say of the definition of "technological protection measure" and of s 116A, as Benjamin Kaplan wrote of the American law even as it stood in 1967, ^[40] that the provisions have a "maddeningly casual prolixity and imprecision". However, in this court no party advanced the proposition that its task on this appeal was satisfied merely by a consideration of the ordinary meaning of the words in the definition of "technological protection measure".

[55] Rather, Sony contended that, unless the term "inhibit" had the meaning given by the Full Court, it was otiose, adding nothing to "prevent". One meaning of "inhibit" indeed is "prevent". However, it may be taken that "inhibit" is used in the definition of "technological protection measure" in one of its weaker senses, while still necessarily attached to an act of infringement. One such sense has been given above with respect to acts of secondary infringement by dealing in an article created by an act of primary infringement. Further, the operation of a copy control mechanism to impair the quality or limit the quantity of a reproduction may be said to hinder the act of infringement. In that regard, there is a legislative antecedent in s 296 of the 1988 UK Act. This, it will be recalled, spoke of devices or means intended "to impair" the quality of copies made. In the present case, the Sony device does not interfere with the making of a perfect copy of Sony's copyright in its computer program or cinematograph film.

Conclusion on construction of definition of "technological protection measure"

[56] The conclusion reached by Sackville J was correct and should not have been disturbed by the Full Court.

[57] There remain for consideration the two grounds advanced by Sony in its notice of contention. To these we now turn.

Sony's notice of contention

[58] By its notice of contention, Sony contends that the decision of the Full Court should be affirmed on grounds in addition to those upon which it succeeded there. Sony has submitted that its device (comprising either or both the boot ROM in the PlayStation console and the access code on PlayStation CD-ROMs) was a "technological protection measure" on three essentially distinct bases. The first which has been dealt with in these reasons was the construction argument concerning "inhibit" and "practical effect".

[59] The second ground was that the device fell within the terms of the definition of "technological protection measure" because it prevented PlayStation users from reproducing in the RAM of an unmodified PlayStation console a substantial part of the particular program embodied in an unauthorised copy of a PlayStation CD-ROM by playing the CD-ROM in that console. This may be called "the reproduction in RAM contention".

[60] The remaining contention was that the device answered the description of the definition because it prevented PlayStation users from making in the RAM of an unmodified PlayStation console a copy of a substantial part of a cinematograph film embodied in an unauthorised copy of a PlayStation CD-ROM by playing the CD-ROM in that console. This may be called "the cinematograph film contention".

[61] Sackville J had rejected all these submissions. In the Full Court, French and Lindgren JJ accepted Sony's argument on the first point but rejected the other grounds. Finkelstein J accepted Sony's arguments on all three grounds. Hence the notice of contention respecting the reproduction in RAM contention and the cinematograph film contention. To these we now turn.

The reproduction in RAM contention

[62] As Lindgren J noted in the Full Court, ^[41] Sony's contention here must be that the protection device prevents or inhibits reproduction of the literary work constituted by the computer program (being the set of statements or instructions embodied in the CD-ROM for a PlayStation game) in a material form in RAM, within the meaning of para (a)(i) of s 31(1) of the Act. ^[42] Section 21 provides that, for the purposes of the Act, reproduction is to be taken to have occurred in various circumstances. One of those is set out in s 21(1A). This states:

For the purposes of this Act, a work is taken to have been reproduced if it is converted into or from a digital or other electronic machine-readable form, and any article embodying the work in such a form is taken to be a reproduction of the work.

[63] The PlayStation console is equipped with random access memory (RAM) which it utilises in order to accelerate its own operation. This it does by copying into its RAM a portion of the computer program stored in the CD-ROM being played. Sackville J noted at least two key features of this process. First, "[t]he RAM's capacity is limited to 2 megabytes", and "[s]ince the game code may consist of up to 580 megabytes ... only a small section of the game code is downloaded and copied". ^[43] Secondly, the "storage [of the copy] in RAM is temporary, in the sense that the data is only stored there until the PlayStation console is shut down". ^[44]

[64] If any conversion of the program from a CD-ROM to RAM answers s 21(1A), there remains, as Lindgren J noted, ^[45] the question whether the reproduction in RAM is "in a material form" within the meaning of para (a)(i) of s 31(1) of the Act.

[65] Sackville J held that a substantial part of the computer program embodied in a PlayStation CD-ROM was temporarily stored in the RAM of a PlayStation console while the game is played: the storage is temporary because the contents of the RAM will be lost if power to the console is disconnected and are displaced as new instructions are downloaded to the RAM. ^[46] That conclusion was not challenged in this appeal. But his Honour held that temporary storage of a substantial part of the computer program did not entail reproduction of it in a "material form". ^[47]

[66] A definition of "material form" was introduced by the Copyright Amendment Act 1984 (Cth) (the 1984 Amendment Act). ^[48] The definition states:

"material form", in relation to a work or an adaptation of a work, includes any form (whether visible or not) of storage from which the work or adaptation, or a substantial part of the work or adaptation, can be reproduced. [Emphasis added]

The definition was introduced to qualify what had been the general understanding that in copyright law a material form was one which could be perceived by the senses. ^[49]

[67] While the 1984 Amendment Act indicates that RAM may constitute a "material form" for the purposes of the Act, in certain circumstances, this is not determinative of this ground in the notice of contention. This is because if the "material form" upon which Sony relies is a form of invisible storage, then this storage must be one from which the work or a substantial part of it "can be reproduced". In effect, as Mr Stevens contends, the legislature amplified the rights of copyright owners with respect to reproduction in invisible forms of storage but did so subject to essential limitations.

[68] Sony submitted that the words of the definition of "material form" after "includes" were not crucial to its success, because as a matter of ordinary language the data stored in the RAM could be said to reproduce the computer program stored in a PlayStation game in a material form. The answer is that given by Sackville J: ^[50] the data were not in a material or corporeal form, but in a non-material, incorporeal form, comprising essentially electronic impulses.

[69] Sony also relied on para 28 of the explanatory memorandum to the Bill for the 1984 Amendment Act introducing the definition of "material form":

The definition of "material form" is new and makes it clear that material form includes such methods of fixation as storage or reproduction on magnetic tape, read only or random access computer memory, magnetic or laser disks, bubble memories and other forms of storage which will doubtless be developed.

As Lindgren J pointed out, that assumes that in some circumstances the electronic impulses stored in RAM are in material form; it does not state a test for distinguishing between the circumstances in which they are and those in which they are not, and it does not say that they are in material form in all circumstances. ^[51]

[70] The closing words of the definition of "material form", namely "can be reproduced", were interpreted by Finkelstein J in his dissenting judgment in the Full Court ^[52] as meaning "may be able to be reproduced". This takes the inquiry concerning materiality of form from the realm of present capability into that of abstract or conjectural possibility. Lindgren J said it was: ^[53]

[170] ... an unrealistic and strained construction to treat the words "can be reproduced" at the end of the definition of "material form" so widely as to encompass "could be reproduced if an additional device, not supplied with the console and not yet available, were to be manufactured and attached to it" or "could be produced if the RAM under consideration formed part of a future modified console".

[71] Earlier, in Australian Video Retailers Association v Warner Home Video Pty Ltd (AVRA), ^[54] Emmett J had appeared to interpret "can be reproduced" as "ordinarily is able to be reproduced". His Honour said ^[55] that "ordinarily it will not be possible to reproduce the contents of RAM in a DVD player". He continued: ^[56]

[103] ... If a DVD player has been modified, such that it is possible to study or use the RAM for the purpose of reproducing its contents, there could be a reproduction of the computer program in a material form within the meaning of s 31(1)(a)(i) of the Act. However, in the ordinary course, temporary storage of a substantial part of the computer program in the RAM of a DVD player will not involve a reproduction of the computer program in a material form.[104] Where a DVD disc is being played by means of a personal computer, it will be possible, where an appropriate additional program is installed in the personal computer, to reproduce the contents of RAM. However, where a computer does not have such a program installed, the use of the computer for the purpose of playing a DVD disc will not involve the reproduction of the computer programs in question in a material form within the meaning of s 31(1)(a)(i) of the Act. [Original emphasis]

[72] The references by Emmett J to what "ordinarily" will not be possible and to what happens "in the ordinary course" explain what is intended in the statutory phrase "can be reproduced". It is not sufficient to consider what might or would result from additional steps such as the use of additional hardware.

[73] With that in mind, it is apparent from the account of the evidence given by Sackville J that Sony's device cannot answer the requirement of the definition of "material form".

[74] Sackville J accepted the evidence of Mr Nabarro, Vice-President, Technical Services, of Sony Computer Entertainment Europe Ltd, as is apparent from the following passage in his Honour's judgment: ^[57]

[51] Once a portion of the game code has been copied into the RAM, it is stored there. The storage in RAM is temporary, in the sense that the data is only stored there until the PlayStation console is shut down. Moreover, as Mr Nabarro explained, the data stored in the RAM will be "flushed" as new instructions are transferred from the PlayStation game's code. Mr Nabarro was also asked whether the portion of the game code stored in the RAM could be extracted and reproduced. His answer was that this could not be done without developing hardware which would enable the process to be reversed. [Emphasis added]

[75] What is said in the last quoted sentence is sufficient answer to Sony's case. It is unnecessary to determine whether the temporary storage which is "flushed" is sufficient to answer the definition of "material form". However, it should be noted that in the formulation of Art 11 of the WIPO Copyright Treaty, to which reference has been made, a proposal was made but not accepted to give explicitly to copyright owners the exclusive right to authorise "direct and indirect reproduction of their works, whether permanent or temporary , in any manner or form". ^[58]

[76] Sackville J observed: ^[59]

[148] On the face of things, it might seem surprising that the reproduction in electronic or digital form of a computer program is not necessarily an infringement of copyright in the computer program. The scheme of the legislation, however, seems to be that reproducing a work in electronic or digital form infringes copyright, pursuant to ss 31(1)(a)(i) and 36(1) of [the Act], only if the form in which the work is reproduced is itself capable of further reproduction ...

His Honour added that this approach is consistent with s 21(1A) of the Act and added: ^[60]

[149] It is plausible that the legislation is structured in this way as a means of balancing the interests of copyright owners and users. If a work such as a computer program is reproduced in electronic or digital form, but is not amenable to further reproduction, it might well be thought too restrictive to regard the first reproduction in electronic or digital form as necessarily an infringement of copyright...

[77] Finkelstein J referred to certain United States authorities holding that the downloading of computer software into the RAM is the making of a "copy" for the purposes of the Copyright Act 1976 (US), as amended, 17 USC §101. That provision defines "copies" as:

material objects ... in which a work is fixed by any method now known or later developed, and from which the work can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device.

The word "fixed" was defined as follows:

A work is "fixed" in a tangible medium of expression when its embodiment in a copy ... is sufficiently permanent or stable to permit it to be perceived, reproduced, or otherwise communicated for a period of more than transitory duration.

[78] Finkelstein J acknowledged that the legislative scheme in the United States is different and the authorities have attracted a great deal of criticism ^[61] . In particular, the utility of the authorities is diluted by the words "with the aid of a machine or device" in the definition of "copies". Lindgren J made the point that the words permit use of a machine or device not already present in the PlayStation console and software of which the RAM is part. ^[62]

[79] The first ground of the notice of contention, the reproduction in RAM contention, should be rejected.

The cinematograph film contention

[80] The term "cinematograph film" is defined in s 10(1) of the Act as follows:

"cinematograph film" means the aggregate of the visual images embodied in an article or thing so as to be capable by the use of that article or thing:

(a)

of being shown as a moving picture; or

(b)

of being embodied in another article or thing by the use of which it can be so shown;

and includes the aggregate of the sounds embodied in a sound-track associated with such visual images.

[81] The set of statements or instructions to be used directly or indirectly in a computer to bring about a certain result answers the definition of "computer program" which, in turn, is brought within the definition of "literary work" within Pt III of the Act. Cinematograph films are differently treated. Cinematograph films are one of those subject-matters other than works in which copyright is conferred by Pt IV of the Act. Section 86(a) provides that, for the purposes of the Act, copyright in relation to a cinematograph film includes the exclusive right "to make a copy of the film".

[82] Section 21(6) states:

For the purposes of this Act, a sound recording or cinematograph film is taken to have been copied if it is converted into or from a digital or other electronic machine-readable form, and any article embodying the recording or film in such a form is taken to be a copy of the recording or film.

Further, s 24 deals with this concept of embodiment in an article as follows:

For the purposes of this Act, sounds or visual images shall be taken to have been embodied in an article or thing if the article or thing has been so treated in relation to those sounds or visual images that those sounds or visual images are capable, with or without the aid of some other device, of being reproduced from the article or thing.

[83] Finally, it follows from s 14(1) that it is sufficient for infringement that there has been the copying of "a substantial part" of a cinematograph film.

[84] Galaxy Electronics Pty Ltd v Sega Enterprises Ltd ^[63] concerned two video games constituting a series of images such that the events represented on the screen varied according to the actions of the player of the game. The Federal Court held that the aggregate of the visual images generated by the playing of each of the games constituted a "cinematograph film". It did not matter that the images were embodied in the computer program or integrated circuits in a different form from that in which they might appear on the screen. Nor did it matter that the images seen by players were created by computer calculations only immediately before their appearance on the screen. The present litigation also concerns games with this general character.

[85] In his evidence at the trial in this case, Mr Nabarro referred to the importance of the interactive elements in the codes for computer games and to the "level" at which particular choices are made to play a game. Earlier, in Galaxy, ^[64] the Full Court adopted what had been said at trial by Burchett J: ^[65]

[E]xcept for the opening and closing sequences, the events represented on the screen will show differences from screening to screening, except where the player's responses are all correct.... the apparatus is designed to screen the simple story only when the correct responses to a series of cues are fed into it by the player; and when incorrect responses are given, a number of variations will result.

[86] Neither side sought in this court to challenge Galaxy , although the amici curiae rightly pointed to difficulties to which that case gives rise. However, what now follows in these reasons proceeds on the footing that the aggregate of the images and sounds stored on a PlayStation CD-ROM answers the statutory description of "cinematograph film".

[87] By its notice of contention, Sony would have this court decide that the RAM of a PlayStation console is an article or thing in which at least a substantial part of a cinematograph film is embodied. It is then submitted that the Sony device was a "technological protection measure". It prevented the making of a copy of a film embodied in the RAM.

[88] This argument was raised at trial by an amendment of pleadings pursuant to leave granted on the first day. The preparation of the affidavit evidence thus had preceded the pleading.

[89] Sackville J's conclusion on this branch of the case was expressed as follows: ^[66]

[160] ... In the absence of clearer and more detailed evidence as to the nature and quality of the images embodied in the instructions stored in the RAM, assessed in relation to the totality of "the aggregate of visual images" constituting the cinematograph film, I cannot conclude (to adopt the language of Emmett J [in AVRA]) that "the ephemeral embodiment" of a small proportion of images in the RAM constitutes the act of making a copy of the cinematograph film for the purposes of s 86(a) of [the Act].

[90] In the result, his Honour decided that, on the evidence before him, the argument founded on s 86(a) of the Act had to be rejected.

[91] The question whether the evidence presented at trial was sufficient to establish the copying of what amounted to a "substantial part" of a cinematograph film which had been copied into the RAM was essentially a matter for Sackville J at trial. Mr Stevens correctly stresses in this court that there was no attempt in the Full Court to assess the "substantiality" of any part of any particular cinematograph film by reference to the whole of it. The determination of questions of what amounts to the taking of a substantial part of a work or other subject-matter is notoriously difficult. This is nonetheless so in the present case where, on the evidence of Mr Nabarro, the purpose of the temporary storage of a small part of the instructions on the RAM of the PlayStation console is purely to enable the display of visual images and sounds by the console in real time.

[92] In his dissenting judgment in the Full Court, Finkelstein J said: ^[67]

[222] ... [T]he question whether a substantial part of the copyright had been reproduced did not require detailed evidence. In particular, it did not require oral evidence from a witness to describe the relevant technology and, perhaps, express an opinion on the issue of substantiality. In many instances, of which this case is a good example, the judge can make an assessment whether a substantial part of the copyright has been taken by making a simple visual comparison between the copyright work and the allegedly infringing work. Second, I do not accept that the judge was confined in his consideration of the issue to what he observed during the demonstration. He had available to him the disks and a PlayStation console and, if necessary, he could have personally played the games in order to assess the nature and the quality of the images stored in the RAM when compared to the totality of the visual images comprising the film.

[93] However, this case did not involve the viewing of a motion picture in the ordinary understanding of that term. The visual images which, consistently with Galaxy , are to be taken to constitute a cinematograph film do not have a set course or sequence of motion. The whole copyright subject-matter cannot be determined merely by a visual appraisal as with a motion picture. The Sony cinematograph films did not comprise visual images and sounds conventionally arranged in a linear sequence. Rather, they were interactive in nature, so that the ability to appreciate directly their content was dependent upon particular activity varying between one player and another.

[94] In this setting, difficult questions for the assessment of substantiality are presented. It is not sufficient here to attempt to assimilate an artificial and contrived demonstration of the playing of the games to the viewing of a segment of a motion picture.

[95] Sony correctly emphasises that the case law concerned with Pt III "works" such as books, where the subject-matter may be appreciated directly, shows that the courts readily enter upon the question of substantiality and that the emphasis has been upon quality, not merely quantity. However, the judgment of Starke J in Blackie & Sons Ltd v Lothian Book Publishing Co Pty Ltd ^[68] is an early illustration in this court of the general proposition that substantiality is a question of degree which depends upon the circumstances of each particular case.

[96] Where the issue of substantiality arises with respect to a computer program, the importance of evidence is apparent from Data Access Corporation v Powerflex Services Pty Ltd. ^[69] In the present case, whether what remained in the RAM, as "accessible by the console", had "an appreciable, playable, coherent, viewable and enjoyable part" of what must be taken to be a cinematograph film was, Sony submits, sufficiently established by the demonstration of two games during the hearing of the Full Court appeal and the reaction thereto of Finkelstein J.

[97] It is unnecessary in this appeal to consider what may be the scope in dealing with computer games for evidence bearing upon the alleged "quality" of that which has been taken. Reference has been made above to the interactive nature of the computer games, and to the limited (and technical) purpose of the temporary storage on the RAM of the PlayStation console. Whatever the scope here for consideration of "quality", there was an unsatisfactory carriage by Sony of its evidential burden. There remains then the question of quantity.

[98] What here is critical is Sackville J's acceptance that the evidence suggested that only a very small proportion of the images and sounds comprising the cinematograph film were "embodied" in the PlayStation console's RAM at any given time. ^[70] In the circumstances as they arose at trial, Sony failed to lay the necessary evidentiary basis for a finding in its favour on substantiality.

[99] That being so, it is unnecessary to consider other submissions put in this court in answer to Sony's case.

[100] The second ground in the notice of contention also fails.

Disposition of the appeal

[101] The appeal should be allowed with costs. The orders of the Full Court of the Federal Court should be set aside. In place of those orders, the appeal to the Full Court should be dismissed with costs.