Contracts by electronic mail

MARCH - APRIL

THE COMPUTER LAW AND SECURITY REPORT

CONTRACTS BY ELECTRONIC MAIL The problem of postal acceptance of contracts was first considered in detail by the courts in 1818.(2) The judge considered three possibilities: either the contract was accepted when the letter was put into the post, or when it was delivered to the offeror's address, or when it was brought to the actual attention of the offeror. He decided on the first since if the contract were not made when the acceptance was put in the post, there would be no reason why the offeree should be bound in contract unless he knew that his acceptance had been received. So he could request confirmation of acceptance. This could lead to an infinite regress, and inevitable delay in commercial transactions. During the nineteenth century, judges reinforced the standing of the postal rules by reference to commonsense and the needs of business. In Harris' Case(3), Mellish LJ referred to "mischievous consequences which would follow in commerce" if no such rule were adopted, while Lord Cottenham stated(4) in Dunlop v Higgins that " commonsense tells us that transactions cannot go on without such a rule." However, the applicability of the postal rules is circumscribed by the restriction that they must not be allowed to result in absurdity and inconvenience, and in Holwell Security Ltd. v Hughes(5) Lawton LJ develops this further: "In my judgement the factors of inconvenience and absurdity are part of a wider principle, namely, that the rule does not apply, if, having regard to the circumstances, including the subject matter under consideration, the negotiating parties cannot have intended that there should be a binding agreement until the party accepting an offer ..... had in fact communicated the acceptance to the other." In 1955, in Entores Ltd. v Miles Far East Corporation(6), Denning LJ refused to extend the postal rules to Telex, on the grounds that it is a form of instantaneous communication. In the same case Birkett, LJ stated that Telex communications "do not differ in principle from the cases where the parties negotiating a contract are actually in the presence of each other," and Parker LJ stated "as a general rule, a binding contract is made at the place where the offeror receives notification of the acceptance, that is, Where the offeror i s " (our emphasis). Since that judgement was delivered, commercial and technological developments have made Telex communication much less "instantaneous" in many circumstances. But as recently as 1982 the House of Lords(7) refused to accept the contention that Entores was wrongly decided, despite the recognition that "since 1955 the use of Telex has been greatly expanded and there are many variants of it," and that "the message may not reach, or be intended to reach, the designated recipient immediately ..... ,,.(8) However, in that judgement, Lord Wilberforce also stated that no universal rule could cover all such cases; and the courts have not yet had occasion to address themselves to those categories of electronic communication facility that are completely new.

C O N T R A C T S BY E L E C T R O N I C M A I L : SOME ISSUES EXPLORED Developments in communications technology have given business organisations new opportunities to interact through electronic media. It is anticipated that the time will come when computer Based Message Systems are regularly used in the negotiation of commercial contracts. Since neither Parliament nor the law courts have yet addressed this matter, it is not possible to give an authorative account of the law on contracts by electronic mail, but we believe it is essential to explore the issues which may arise out of the application of this new technology to business communications. Contract law is not an area in which Parliament tends to intervene and the courts cannot decide the issue until a dispute is brought before them. However until such time as they have cause and opportunity to address the issue, users and providers of electronic mail should attend to the legal dimension. It is hoped that the ensuing discussion will be helpful to users, service-providers and their legal advisers.

Resume of contract law Before a contract can be said to exist "there must be some external manifestation of assent, some word spoken or act done by the offeree or his agent which the law can regard as the communication of the acceptance to the offeror"(1). Thus the general rule in English law is that an acceptance is only effective when communicated to the offeror. One problem which may arise when parties make contracts by communicating over some distance is that the relevant circumstances may change before a reply to the offer is received: the offeror may then wish to withdraw his offer. What is the legal position if his letter revoking the offer should cross in the post with the other party's acceptance? This uncertainty is avoided by the doctrine that the contract is made when the letter of acceptance is put into the post. Since the offeror can specify the mode of acceptance, he could exclude the post and thereby eliminate the risk that, unknown to the offeree, the letter might not arrive. Since it is for his own benefit that the offeror opts to deal by post, it is considered that he thereby waives the normal rule that the contract is not made until he actually receives the acceptance. The acceptance is treated as constructively communicated to him when it is posted, although not in fact communicated to him until he receives it. This exception only applies in restricted circumstances. Either it must be clear that the offeror has expressly or by implication prescribed that the post is an acceptable mode of communication or, if no mode of acceptance has been specified by the offeror, the nature of the offer and the circumstances in which it was made must be such that postal acceptance was appropriate. Relevant circumstances will be taken into account in deciding whether a particular communication mode has been specified. These might include a past course of dealings, implying a willingness to be bound by the same mode of acceptance in the future as was regularly used in the past. Alternatively the mode of communication used to make the particular offer may carry an implication about the mode that should be used for its acceptance. Thus an offer sent by telegram would have been taken as evidence of a request for a speedy reply. If no mode is specified by the offeror then the form of communication to be used for their reply depends on the nature of the offer and the circumstances in which it was made.

Relevant features of electronic message systems We use the term Electronic Message Systems (EMS) to include the full range of communication systems. Thus Computer Mail Systems (CMS) and Telex, for example, are both types of EMS. We divide CMS into Single-Host Mail Sytstems (SHMS) and Network Mail Systems (NMS). Where we refer simply to CMS, the reference should be taken to cover both Single-Host and Network variants. 2

THE C O M P U T E R LAW A N D S E C U R I T Y R E P O R T

EMS, and their conventional forerunners, can be classified according to both technical and social features. Many features may be quite incidental for present purposes; the technical features that appear potentially important are Addressing Identity and Delay. The relevant social variables are Ownership, Control and Regulation~ By Delay, we mean the possibility of delayed communication. Non-delayed systems are those without storage capacity, which only support instantaneous communication. Conventional Telex and conventional telephone are nondelayed, and so is direct terminal-to-terminal working on a multi-access computer system. Store-and-forward Telex, telephone answering systems and CMS (whether Text or Voice), are delayed. It should be noted that a delayed EMS may also allow instantaneous working. On the other hand, conventional post has a built-in inevitability of delay, which differentiates it from any of the electronic media. The Identity Addressed may in broad terms be either a Machine (at a place) or a User (or Group). Where communication is directed to a User Identity, this may either be to an electronic mailbox (delayed communication to a user identity) or directly to the terminal where he is logged in (instantaneous communication to a user identity - a facility unlikely to be used for commercial messages). Many NMS are hybrid, in that messages will be addressed to a user identity within a given host machine, but where a NMS provides an automatic routing directory to switch messages to the correct host computer, then addressing may be simply to User identity. Ownership and Control of equipment, files and messages may turn out to be relevant to the interpretation of communication events, and to the obligations of parties to a negotiation. Regulations imposed by public policy upon the providers of communication systems might also have a bearing. It should be noted that postal rules do not apply to courier services. An EMS provided under a VANS licence appears to have the characteristics of a courier service rather than a postal service. In general, the variables of Delay and Addressing Identity appear the most crucial for the problem-families we discuss below; Ownership and Control could be important in some circumstances, moderated by the technical variables, while Regulation would have at most a minor role to play. EMS can offer a great variety of services with many combinations and permutations of technical features, and it is not necessarily obvious how these characteristics should be related to the law developed to deal with existing modes of communication. In broad terms, we may ask: should text message systems be assimilated to conventional mail systems, to courier services or to Telex? Should sending a message via a voice mail system be treated differently from telephoning and leaving a message on an answering machine? Or is it necessary to invoke more fundamental principles to clarify the implications of completely new modes of communication? At a more detailed level, we may identify the following families of problems: 1. Problems relating to the concepts of "posting", "delivery" and "receipt" in EMS. 2. Problems of applying the concept of "place of occurrence" to electronic communication events, given the possibility of "remote access" to mail-boxes by both parties. 3. Problems arising from the difficulty of applying the concept of "instantaneous communication" as message handling becomes increasingly sophisticated.

6 CLSR

4. Problems relating to ownership and control over the message after it has been "posted." 5. Problems relating to possible consequences of using a non-human "agent" in handling and filtering messages, including the use of "active messages" to generate answering messages. Our discussion concentrates on the first three areas, although we would not wish to underestimate the potential importance of the others. We shall discuss these problems in relation to a classification of relevant features of EMS. "Posting", " D e l i v e r y " and " R e c e i p t " in CMS If posting, delivery and receipt are the same event in a CMS, then it becomes a form of instantaneous communication, and the postal rules could not be applied; but if there are in principle two or three separate events, then it is non-instantaneous. A CMS provides a "mailbox" for each user. There is some inconsistency of terminology, with "mailbox" being used by some authorities to refer to a store of messages, and by others to refer to a program which gives users access to these messages. We shall use the term "mailbox" to refer to a file or directory of incoming messages. The sender creates a file, possibly temporary, containing the message; then he gives a command to "send" this message to another user. On SHMS, it is then immediately added to the recipient's mailbox. On NMS the message may be queued for some time before being transmitted and may indeed be queued more than once in the course of its transmission. It seems clear enough that "posting" occurs when the sender gives the command to send his message. However, unlike conventional mail, there may be neither legal barrier nor practical obstacle to the sender deleting his message before the addressee has had a chance to read it. Thus the action of "posting" in a CMS may not commit an offeree to acceptance so irrevocably as posting a letter would do. It is possible to make posting irreversible, but this may have unintended negative consequences, which we discuss below. "Posting" in CMS is at least a readily identifiable event, but "delivery" and "receipt" are much more elusive. A user can decide when he wants to "collect" the messages in his mailbox. To do so he must be logged in to the host computer, possibly through the Public Switched Telephone Network and/or through a public or private computer network. Regular users of CMS are likely to log in for many tasks other than sending and receiving mail, since for higher-grade business users systems that only support the mail function tend to be unsuitable even as a communication device.(9) CMS may offer several options when reading mail, e.g. skimming headings, selection by subject or sender, forwarding, delegation. The range of options offered is by no means uniform, and sender and receiver may handle their mail through different US programs, offering different facilities. The concept of delivery is also related to responsibility for ensuring successful communication. Where the postal rules do not apply, the offeree is responsible for seeing that his acceptance is actually communicated to the offeror. However, the offeror could not rely on this in instantaneous communication if he knew that the offeree had attempted to communicate with him, that the attempt had been unsuccessful and that the offeree was not in a position to know that his attempt to communicate had failed, yet he, the offeror, did not notify the offeree of the failure (see below). Facilities for checking the status of users and of messages may be relevant here.

MARCH

-

APRIL

THE C O M P U T E R LAW A N D S E C U R I T Y REPORT

On most CMS the user will be notified of any mail awaiting him when he logs in, and if he is already logged in when a message "arrives" he will be notified of its "arrival" by a prompt, e.g. "YOU HAVE MAIL." This prompt may be displayed immediately, or it may not appear until there is a suitable break in the flow of work. On a SHMS, when the sender gives the command to send his message, it becomes almost instantaneously available to the recipient to read. However, if the recipient is not logged in he will not be aware that an attempt has been made to communicate with him. In this respect, the sender is in a similar position to the user of conventional mail: he trusts to the reliability of the postal service, expecting that his letter will be delivered. However, many CMS will also provide "status" information that will enable him to check whether his message has been read, and/or when the addressee last logged in, etc. Some systems also give a warning if one sends a message to a user who has not been logged in recently. Does "delivery," then occur when the message is added to a user's mailbox, or when he is notified of its "arrival" or not until he causes the message itself actually to be displayed for him on the VDU, or to be printed out as "hard copy"? On the first interpretation, posting and delivery are simultaneous in a SHMS, and so communication is in a sense "instantaneous" but the situation differs considerably from that envisaged by Denning LJ in his judgement establishing the non-applicability of the postal rules to Telex.(10) Lord Denning discusses the possible effects of technical or human failure leading to the non-receipt of the message. He discusses both line failures (which, he says, the offeror can recognise by the stopping of the teleprinter motor) and failures in the receiving apparatus which can be reported by the user at the receiving end. He goes on: "In all the instances I have taken so far, the man who sends the message of acceptance knows that it has not been received, or he has reason to know it. So he must repeat it. But suppose that he does not know that his message did not get home. He thinks it has. This may happen if the listener on the telephone does not catch the words of acceptance, but nevertheless does not trouble to ask for them to be repeated: or if the ink on the teleprinter fails at the receiving end, but the clerk does not ask for the message to be repeated: so that the man who sends an acceptance reasonably believes that his message has been received. The offeror in such circumstances is clearly bound, because he will be stopped from saying that he did not receive the message of acceptance. It is his own fault that he did not get it. But if there should be a case where the offeror without any fault on his part does not receive the message of acceptance, yet the sender of it reasonably befieves it has got home, when it has not, then I think there is no contract." In Brinkibon, Lord Fraser of Tullybelton also states that "once a message has been received on the offeror's Telex machine, it is not unreasonable to treat it as delivered to the principal offeror, because it is his responsibility to arrange for prompt handling of messages within his own office."(11) With the office becoming "less of a place, and more of a system, ,,(~2) one must consider where the limits of this responsibility might now be seen to lie. Is the mailbox or the VDU the equivalent of the "Telex machine"? Several combinations of circumstances could be envisaged, peculiar to CMS, in which there could be argument about contract, and about responsibility for successful communication. Features intended to guard against fraud or deceit, or to prevent misunderstanding, might in fact have

negative effects. For example, inability to delete a message once sent might seem to provide a level of security, but could actually lead to misunderstanding and uncertainty; status information appear to provide an element of reassurance, but could on occasion mislead etc. We illustrate these possibilities by reference to three hypothetical scenarios. a) ABC uses a CMS to communicate an offer to XYZ and later changes his mind. Since this particular CMS does not allow ABC to recall or alter his message once sent, he sends a second message revoking his offer. XYZ is "off line" at the time the two messages are sent. He logs in next day and receives ABC's offer to which he immediately replies with a message of acceptance. He then reads the next item of mail, which turns out to be ABC's second message, rescinding the offer. Is there a contract? Here the relevant rule is that the offeree has to know of the revocation, before it becomes effective. So the revocation was not effective when XYZ sent his message of acceptance. But there is uncertainty arising from the problem: when is ABC considered to have received XYZ's message of acceptance? If this is when the message is delivered at ABC's mailbox (ie instantaneous communication), then the offer was actually accepted (acceptance was communicated) before XYZ knew of the revocation: so there is a contract. But if the acceptance is not considered to have been communicated until ABC reads it, then XYZ may know of the revocation before his acceptance becomes effective, so XYZ does not know if the contract exists until he has checked the status of the acceptance message, and seen that ABC or his agent had read it before he read ABC's message. b) ABC has made an offer to XYZ, who sees the message on his VDU at a time when both he and ABC are logged in. However, at the very time that XYZ is composing his letter of acceptance, ABC is composing a letter rescinding the offer. Both companies give the command to send their letters at the very same instant. Is there a contract? All the complexities of scenario (a) apply here; but in addition it could be held that there was no true consensus, and therefore no contract, or else that the balance of doubt should be in favour of the offeree since the offeror had implicitly taken on the risk of such an occurrence by using a CMS for contract negotiation. A case of this type might lead to records of systems status becoming relevant evidence for the existence of the contract, particularly if fine matters of message timing were in question. c) ABC offers to sell some goods to XYZ, and the latter sends a message of acceptance, both by CMS. ABC reads his mail and finds that XYZ's message is unintelligible, owing to data corruption. The effect of ABC reading his mail is, of course, that the systems records XYZ's message to ABC as having been read. XYZ seeks status information from the computer, and seeing that ABC has now read his letter of acceptance judges that the contract is now made; he therefore immediately re-sells the goods to PQR and goes out to celebrate, even while ABC is composing a letter to XYZ informing him that his (XYZ's) letter was unintelligible. Before XYZ had read ABC's second letter ABC sells the goods to someone else and sends XYZ a third letter cancelling the offer. In this case, the offeror would appear to bear the risk, since he has allowed the medium to be used. However, in the case of a NMS, Ownership and Control might be relevant. If each party owner his own host computer, there might be scope for ABC to argue that the fault lay with, for example, poor hardware or software maintenance on XYZ's host computer.

4

T H E C O M P U T E R LAW A N D S E C U R I T Y R E P O R T

6 CLSR

However, poor network maintenance would be at the offeror's risk. Any user who had a program handle and acknowledge messages he had not actually read, thus generating misleading status information, would be liable for the consequences.

If the postal rules are extended to cover electronic mail then the contract will be made once the command is given to send the acceptance. This would be so, even though the element of inevitable delay which gave rise to the postal rule could well be absent, thus ordinary non-delayed telex and store and forward telex (or systems with similar facilities) would be treated differently. In Brinkibon, Lord Brandon said "The cases on acceptance by letter and telegram constitute an exception to the genera/ principle .... The reaon for the exception is commercial expediency .... That reason .... applies to cases where there is bound to be a substantial interval between the time when the acceptance is sent and the time when it is received."(16, our emphasis). The fact that instantaneous communication is possible by CMS could be taken as grounds for refusing to extend the postal rules to cover them. The problem here would centre on receipt. The place of receipt must be the offeror's mailbox. Otherwise the offeree could be dependent on the offeror's travel schedule if he uses the remote access facility to read his mail. If the 'normal' rules are applied that is those applicable to telex then the contract will be made at the time and in the place, that the acceptance is received. It is submitted that although no universal rule can cover all cases the courts would be reluctant to extend the postal rules. The responsibility must lie with users and providers of EMS, to arrive at an appropriate definition of "commercial reality." The legal rules were developed to deal with situations that had been left undefined by the parties. But it is always open to the offeror to specify the mode of acceptance of the offer, and he can do so in sufficient detail to avoid the ambiguities concerning time and place of receipt that we have identified above. Likewise, the proper law of the contract can be explicitly stated in the contract. Until normal commercial practice has developed a system of mutual understanding of the way in which these issues should be handled, parties to negotiation should spell out explicitly in the contracts the intended answers to these questions. The use of standard documents in preparing messages should ease the task of ensuring that this is done; alternatively, service providers could issue a set of standards which they require every user to accept, unless he gives specific notice to the other party that he intends otherwise.

2. The concept of "place"

The "place" where the contract is made becomes important in deciding the proper law of a contract made between parties in different countries, and in deciding whether an action may be brought in the English courts where the defendant is outside the jurisdiction. (13) It is thus possible to envisage the situation where two communicators are in two different jurisdictions, and the computer through which they are communicating is in a third jurisdiction. It is the human beings, not the computer, who are parties to the contract; but the place where the contract is made may depend on which definition of "delivery" prevails. If arrival at the mailbox constitutes delivery, then the acceptance is received by the offeror at a time when he is not physically present. It communication of acceptance is taken to be when the message is actually read, then the place of the contract, and hence the law of the contract, may be uncertain to the offeree, which is unsatisfactory. 3. " I n s t a n t a n e o u s sophistication

Communication"

and

system

Computers and microcomputers may be used in EMS which are not true CMS. In particular, a computer may be used either as a Switching Device or as a Message Preparation Device. Use of Telex Switches and Store and Forward exchanges implies that a Telex message can be sent but not delivered instantaneously because of the receiver's machine being busy. The increasing sophistication of the Telex system makes it less and less resemble direct instantaneous person-to-person communication. There are also plans for Telex to interwork with Teletex, which is not only itself a "super-telex," but will also be used as a transport system for NMS. Thus a message might originate within one type of system, but end up in a different system, throwing further doubt upon the assumptions the communicators might reasonably make. This adds further complexity to the situation described by Lord Wilberfome in Brinkibon: "The sender and recipients may not be the principals to the contemplated contract. They may be servants or agents with limited authority. The message may not reach, or be intended to reach, the designated recipient immediately: messages may be sent out of office hours, or at night, with the intention, or on the assumption, that they will be read at a later time .... The messages may have been sent and/or received through machines operated by third persons .... ,,.(14) The House of Lords judgement in Bdnkipon does not however necessarily imply a rigid or narrow view as may be seen from Lord Wilberforce's further remarks "No universal rule can cover all such cases; they must be resolved by reference to the intentions of the parties, by sound business practice and in some cases, by a judgement where the risks should lie."

Sharon Harvey Gloucestershire College of Arts and Technology Julian Newman Report Correspondent

City of London Polytechnic. © North Holland 1984 (a division of Elsevier Science Publishers). This is a revised version of a paper originally presented at the International Working Conference on Computer Based Message Services held at Nottingham on 1-4 May 1984. Notes and References

(1) Cheshire& Fifoot's Law of Contracts (111hedition, 1986). London: Butterworth. (2) Adams v Linsell, (1818) 1 B & Aid 681 (3) Harris' case: re Imperial Land Co. of Marseilles (1872) LR7 CH APP 587 (4) Dunlop v Higgins, (1848) 1 HLC 381 (5) Holwell Securities Ltd v Hughes [19741 1 All ER 161, [19741 1 WLR 155 (6) Entores, Ltd v Miles Far East Corpn, [19551 2QB 327, I1955] 2 All ER 493 (7) Brinkibon Ltd v Stshlag Stahl und Stahlwahrenhandels gesellschaft [1982] 1 All ER 293 (8) Lord Wilberforce in Brinkibon (see note 7)

Some solutions

It is anticipated that the courts will at some future time, be called upon to decide the position relating to acceptances by electronic mail.(is) In so doing they would wish to avoid both uncertainty about the place and time of contract and a decision against commercial reality or commercial expediency.

5

MARCH -

THE COMPUTER LAW A N D SECURITY REPORT

APRIL

(9) Newman,JC (1981) "Human factors for CBMS for Management Use," in Uhlig, R (ed) Computer MessageSystems.Amsterdam: North Holland. (10) Denning LJ (Ioc. cit.) (11) Lord Fraser of Tullybelton in Brinkibon (see note 7) (12) Tapscott, D (1982) Office Automation: A User-Driven method. New York, NY: Plenum. (13) Serviceof a writ outside the jurisdiction may be allowed, at the court's discretion, under RSC Order 11, if the contract was made

within the jurisdiction. Howeversince the Civil Jurisdiction and Judgements Act 1982 this will not apply in an action brought against a person domiciled in another EEC country (14) See note 7. (15) So far the courts have only commented on the unreliability of the system. Cornhill Insurance PLC v Improvement Services Ltd (1986) 1 WLR 114, 116 (16) Lord Brandon in Brinkibon (see note 7)

PRIVACY LESSONS FROM THE AUSTRALIA CARD DEUX EX MACHINA? Readers of The Report will be familiar with the Australian Labor Government's proposals for a national identification scheme, the psuedo-patriotically named 'Australia Card', and many of the arguments against it, from Roger Brown's recent article1, In my view, it would have gone beyond being a mere identification scheme, and would have established the most powerful location system in Australia, and a prototype data surveillance system totally out of keeping with the traditions of a common law country 2. It is no exaggeration to say that the ID card proposal gave many Australians nightmares, and not only those of East European and Asian extraction. However, like all good nightmares, we woke one September morning to find that the nightmare was over. A retired and retiring public servant, Ewart Smith, himself having a restless night being kept awake by magpies, had dawn upon him a loophole in the legislation. It ensured that the Opposition parties could prevent the Card from ever being implemented, even though the Government could still pass the Australia Card Bill 1986. For those not familiar with Australia's constitutional structure, this requires explanation.

A thrice denied Bill The background to the Bill's demise is that had been twice passed by the House of Representatives (the lower house of Australia's Federal Parliament) and twice rejected by the Senate (the upper house), in which the Opposition parties held a majority. Australia's Constitution (s57) provides that deadlocks between the houses may be resolved by the Government calling an election in which both houses are dissolved. After only the sixth 'double dissolution' election in Australia's history, called on the pretext of the Australia Card Bill's rejection, the Labor Government was returned to office in June 1987 and vowed to press ahead with the Bill. The Constitution provides that, if the Bill is rejected for a third time by the newly-constituted Senate, a Joint Sitting of both Houses may be called, at which a majority may pass the Bill into law. It is an uncharted constitutional minefield.

The 'loophole' Ewart Smith's 'loophole' depended upon the fact that the Bill

did not explicitly make it legally compulsory for any person to obtain a Card, because this would be of dubious constitutional validity. It merely made it impossible to operate in Australian society without one. However, the crucial clauses of the Bill, those which imposed $20,000 penalties on businesses which failed to require a person to produce their ID Card, or authorised the freezing of bank account and social security payments for those who did not produce one 3, were all defined to commence operation 'on or after the first relevant day'. This 'first relevent day' was defined (cl 32) such that it could only be declared by regulations, and the Government did not have the necessary majority in the Senate to prevent

the Opposition parties disallowing any regulations which were made. The Opposition vowed that no 'first relevent day' would ever occur and that the Bill was 'stone dead'. In effect, if the Bill was passed, obtaining a Card would really be voluntary because no adverse consequences would attach to failure to hold one. The Government was hoist on its own petard.

The untried 'escape mutes' The Government could still have pressed ahead with the legislation, to attempt to get it onto the Statute books and force the Opposition actually to go ahead with what Labor claimed was 'constitutional vandalism'. It is also possible that this loophole was not fatal to the Bill, despite the Opposition's claims. Among the Government's possible 'escape routes', it has been argued 4 that the disallowance of regulations would not serve to alter a 'first relevant day' once it was declared; that a 'first relevant day' could be proclaimed, the provision for regulations being only directory, not mandatory; that the Bill could have been amended to allow a proclamation, not regulations, at the joint sitting of Federal Parliament necessary to pass the Bill; and that the blockage of the Bill by disallowance of regulations is unconstitutional, being contrary to the intent of the provisions of s57 of the Constitution. AI! of these potential escape routes were of dubious validity and would inevitably have resulted in lengthy and bitter High Court and Parliamentary challenges. The Government considered its options for a week and then abandoned the Bill, blaming the Opposition's 'constitutional vandalism', and attempting to draw parallels with the sacking of the Whittam Labor Government in 1975.

Political realities and public opinion What lessons should those who are interested in data protection and the politics of surveillance draw from this? Was it merely a lucky escape for the lucky country, a deus ex machina for which we should all toast Ewart Smith and an unknown Parliamentary draftsman who preferred regulations to proclamations? If we only consider the mechanics of the Card's demise, this is how it may appear, but this would be to completely misread the real reasons for the abandonment of the Card and to miss the important lessons which should be drawn from the Australian experience. Since the idea of an ID card to combat tax and welfare fraud was first floated in mid-19855, public opinion polls indicated consistent majority support for the Card of about 680, up to the time of the June 1987 election 6, The Labor Government was returned to office at an election ostensibly called because of the Senate's rejection of the Bill. In the election campaign the Card was rarely mentioned by either side, the principal issues being economic ones. There was certainly no public outcry against the Card, and no apparent electoral damage to Labor. Yet a mere three months later the Bill was dropped, ostensibly because of a legal loophole, but one which was possibly not fatal (as has been mentioned). There was no suggestion that the Government would introduce a fresh