Your Response Ltd v Datateam Business Media Ltd

This is an appeal against the order of District Judge Bell made following the trial of a claim by the respondent, Your Response Ltd, against the appellant, Datateam Business Media Ltd, for sums alleged to be due under a contract for the management of an electronic database and for damages for breach of contract. By his order the judge gave judgment for the respondent on its claim and dismissed the appellant’s counterclaim for damages. An appeal from the order of the District Judge lies to this court because the order is a final order made in a Part 7 claim allocated to the multitrack. It raises an interesting and important question, namely, whether it is possible to exercise a common law possessory lien over an electronic database.

The facts giving rise to the dispute between the parties can be described quite shortly. The appellant (“the publisher”) publishes a number of magazines which are distributed to a large number of different subscribers. (Although some of the publications are made available to readers at no charge, being paid for by advertising revenue, it is convenient for present purposes to refer to the recipients generally as subscribers.) For the purpose of managing printing and distribution and for other purposes relating to its business the publisher keeps records containing information relating to subscribers, such as their names, addresses, the publications they receive and other information necessary to enable it to operate its business efficiently. By the end of 2009 the information was being held in electronic form in what is commonly known as a database. It required regular amendment to ensure that the information it contained was up-to-date. The evidence suggests that hundreds of amendments might be required in a single day.

The respondent (“the data manager”) carries on business as a database manager, that is, it offers customers the service of holding electronic databases and amending them as necessary in order to ensure that the information they contain is up-to-date. In about March 2010 the publisher engaged the data manager to hold and maintain its database of subscribers. The contract was not embodied in a formal agreement, but was made partly orally and partly in writing. It was agreed at trial, however, that the best evidence of the main terms agreed between the two parties was to be found in an email sent by the data manager to the publisher on 26^th February 2010. Unfortunately, that email was silent about the means by which the database was to be transferred to the data manager or what was to be done with it when the contract came to an end. Indeed, the email was silent about the circumstances in which, or the terms on which, the contract could be terminated.

During the summer of 2011 things started to go wrong. The publisher was not happy with the service the data manager was providing and privately resolved to bring the contract to an end. On 17^th October 2011 there was a telephone conversation between Mr. Alex Whetton, one of the publisher’s staff, and Mr. David Hooker, the proprietor of the data manager, in the course of which Mr. Whetton purported to terminate the contract with effect from the end of the month. The date of termination was subsequently extended to 16^th November, representing a total of one month’s notice, which the publisher considered sufficient.

On 31^st October 2011 the data manager sent the publisher an invoice for fees due at the end of August 2011. Under the contract payment was due at 30 days. The next day (1^st November 2011) Mr. Whetton asked the data manager to provide certain data and the two of them spoke on the telephone the day after (2^nd November 2011). It seems that their exchanges did nothing to resolve their differences, because the data manager failed to provide the data that Mr. Whetton had requested and on 9^th November 2011 it stopped providing any services under the contract. Thereafter there was an impasse: the data manager refused to release the database or give the publisher access to it until all outstanding fees were paid; the publisher refused to pay until the database was made available to it. In the event the publisher engaged another company to reconstitute the database.

On 15^th February 2012 the data manager started proceedings claiming fees alleged to be due for work carried out under the contract between 1^st August and 9^th November 2011 and damages for repudiation by the publisher of the contract. The publisher counterclaimed for damages for breach of contract represented by the cost of reconstituting the database.

Although a number of questions were argued before the judge, only two arise for consideration on this appeal: (i) what period of notice the publisher was required to give in order to bring the contract to an end; and (ii) whether the data manager was entitled to exercise a lien over the database pending payment of its outstanding fees. In an impressive and carefully reasoned judgment the judge held that a reasonable period of notice was three months, as the data manager had contended, and that the publisher had therefore repudiated the contract by making it clear that it would not continue to perform after 16^th November 2011. He also held that the data manager was entitled to withhold the data until its outstanding fees were paid. He rejected the publisher’s argument that the exercise of a lien was inconsistent with the terms of the contract. He also rejected the argument that it is not possible to exercise a lien over intangible property, in this case electronic data.

Mr. Susman Q.C. for the publisher and Mr. Cogley Q.C. for the data manager both concentrated their attention almost entirely on the second issue, which undoubtedly raises some interesting and difficult questions. It is also a matter of some general importance and I therefore propose to consider it first.

The judge was clearly aware of the need for the law to keep abreast of technological developments and he appears to have been struck by the analogy that can be drawn between information kept in hard copy in the form of ledgers (over which a book keeper could exercise control by means of physical possession) and information kept in electronic form, over which a data manager could exercise control by electronic means. Thus in paragraph 111 of his judgment he said:

“It seems to me in the present case that a lien can apply to the electronic data which was in the possession of the Claimant. It would not be appropriate for the law to ignore the development in the real world of record keeping moving from hard copy records into electronic media. The decision which I have to reach today is of limited purview and no doubt this topic may arise again in other cases in other contexts. But for the purpose of the particular decision which I have to reach in this case, I do not accept the submissions by counsel for the Defendant that a lien cannot exist over the electronic data which was in the Claimant’s possession in just the same way as it could exist over the hard copy records in the Claimant’s possession.”

I have some sympathy for that view, but the judge did not have his attention drawn to some of the leading authorities, in particular OBG Ltd v Allan [2007] UKHL 21, [2008] 1 A.C. 1, and in any event I think it is necessary to examine whether it is consistent with principle.

In this case the court is concerned only with the exercise of a common law lien, which, in its origin, permits a bailee in possession of chattels to refuse to redeliver them to the bailor until he has received payment of outstanding sums due to him. In Tappenden v Artus [1964] 2 Q.B. 185 the owner of a van allowed a customer to use it pending completion of a hire-purchase agreement. The van broke down and was delivered to the defendant for repairs. The price of the repairs remained outstanding and a question arose whether the garage could exercise a lien over it against the owner. At page 194-1995 Diplock L.J. described the common law possessory lien in the following terms:

“The common law lien of an artificer is of very ancient origin, dating from a time when remedies by action upon contracts not under seal were still at an early and imperfect stage of development: see the old authorities cited by Lord Ellenborough C.J. in Chase v. Westmore (1816) 5 M. & S. 180. Because it arises in consequence of a contract, it is tempting to a twentieth-century lawyer to think of a common law lien as possessing the characteristics of a contractual right, express or implied, created by mutual agreement between the parties to the contract. But this would be to mistake its legal nature. Like a right of action for damages, it is a remedy for breach of contract which the common law confers upon an artificer to whom the possession of goods is lawfully given for the purpose of his doing work upon them in consideration of a money payment. If, pursuant to the contract, the artificer does his work, he is entitled to retain possession of the goods so long as his charges, whether agreed in advance or (if not so agreed) payable upon a quantum meruit, are satisfied. The remedy can be excluded by the terms of the contract made with the artificer either expressly or by necessary implication from other terms which are inconsistent with the exercise of a possessory lien; cf. Forth v. Simpson (1849) 13 Q.B. 680, in the same way as the common law remedy in damages for breach of contract may be excluded or modified by the terms of the contract itself. But this does not mean that the remedy of lien, any more than the remedy in damages, is the result of an implied term in the contract to which what we may conveniently call the Moorcock (1889) 14 P.D. 64 criteria relevant to implying terms in a contract apply. The test whether or not the remedy exists is not whether or not its existence is necessary to give business efficacy to the contract. Judged by this test there would in modern times never be an artificer's lien.

The common law remedy of a possessory lien, like other primitive remedies such as abatement of nuisance, self-defence or ejection of trespassers to land, is one of self-help. It is a remedy in rem exercisable upon the goods, and its exercise requires no intervention by the courts, for it is exercisable only by an artificer who has actual possession of the goods subject to the lien. Since, however, the remedy is the exercise of a right to continue an existing actual possession of the goods, it necessarily involves a right of possession adverse to the right of the person who, but for the lien, would be entitled to immediate possession of the goods. A common law lien, although not enforceable by action, thus affords a defence to an action for recovery of the goods by a person who, but for the lien, would be entitled to immediate possession.

Since a common law lien is a right to continue an existing actual possession of goods (that is to say, to refuse to put an end to a bailment) it can only be exercised by an artificer if his possession was lawful at the time at which the lien first attached.”

The emphasis in this description on the “actual possession of goods” makes it necessary to consider a number of questions: what at common law is understood by actual possession, whether it is possible to have actual possession of an intangible thing, whether it is open to this court to recognise the existence of a possessory lien over intangible property and if so, whether it would be right for it to do so.

In the protection of rights of personal property the common law historically drew a distinction between tangible and intangible property. Tangible property, usually referred to as chattels but sometimes as choses in possession, could be the subject of physical possession and thereby physical control, whereas intangible property, consisting of rights to benefits obtainable only by action (and thus known as choses in action), could not. Thus, in Torkington v Magee [1902] 427, Channell J. was able to describe a “chose in action” as meaning “all personal rights of property which can only be claimed or enforced by action and not by taking physical possession.” The view hitherto established has been that the common law does not recognise any third kind of property: see Colonial Bank v Whinney (1885) 30 Ch. D. 261, (1886) 11 App. Cas. 426, to which I shall refer in more detail later in this judgment.

In OBG v Allan receivers invalidly appointed by the claimant’s creditors terminated by negotiation a number of contracts between the claimant and various sub-contractors. The question arose whether wrongful interference with contractual rights could constitute the tort of conversion. The House of Lords held that it could not, because the tort of conversion applies only to chattels and not to choses in action. Lord Hoffmann, who gave the leading speech of the majority, pointed out that historically conversion was a tort against a person’s interest in a chattel and expressed the view (paragraph 97) that the whole of the statutory modification of the law of conversion had proceeded on the assumption that it applies only to chattels. He also noted that in its 18th Report (Conversion and Detinue) in 1971 (Cmnd 4774) the Law Revision Committee had treated detinue and conversion as confined to wrongful interference with chattels, an assumption reflected in the Torts (Interference with Goods) Act 1977, which defines wrongful interference with goods as including “conversion of goods” (section 1) and “goods” as including “all chattels personal other than things in action and money”. Lord Brown of Eaton-under-Heywood rejected the suggestion of Lord Nicholls and Lady Hale that the tort of conversion should be extended to cover the appropriation of things in action partly on the grounds that it would sever the link between the tort of conversion and the wrongful taking of physical possession of property.

As OBG v Allan makes clear, the essence of conversion is a wrongful interference with the possession of tangible property. For these purposes the common law draws a sharp distinction between tangible and intangible property. Even in the case of the conversion of valuable documents (cheques etc.), to which several of their Lordships referred, there is an unlawful interference with a physical object to which a commercial value can be attached. In contrast to chattels, choses in action are intangible things and incapable of the physical possession necessary to support a claim for conversion.

As Diplock L.J. observed in Tappenden v Artus, the essential nature of a common law artificer’s lien (which is the nearest analogy to the lien which the data manager sought to exercise in the present case) is a right to retain possession of goods delivered to him for the purpose of carrying out work on them. Although the right to exercise such a lien has been recognised in a wide variety of cases (see Halsbury’s Laws of England, 5^th edition. volume 68, paragraph 845), the respondent was unable to identify any case in which a right to exercise a lien over intangible property has been recognised. The reason is not difficult to find: whereas it is possible to transfer physical possession of tangible property by simple delivery, it is not possible to deal with intangible property in the same way. Although it is now possible by virtue of statutory provisions to transfer the legal title to choses in action, it is not possible to transfer possession of them in any physical sense. (I ignore for these purposes negotiable instruments and other documentary securities which take a physical form and are thus capable of being converted, their value being treated as the value of the obligation which they embody.) Indeed, I do not think that the concept of possession in the hitherto accepted sense has any meaning in relation to intangible property.

In addition there are indications elsewhere that information of the kind that makes up a database (usually, but not necessarily, maintained in electronic form), if it constitutes property at all, does not constitute property of a kind that is susceptible of possession or of being the subject of the tort of conversion. The Copyright and Rights in Databases Regulations 1997 implement the provisions of European Council Directive No. 96/9/EC. Their primary purpose is to grant protection to the authors of databases by extending to them the protection accorded to literary works by the Copyright, Designs and Patents Act 1988 (“the 1988 Act”). A database for this purpose includes a collection of data arranged in a systematic or methodical way which are individually accessible by electronic or other means (section 3A of the 1988 Act, as inserted by Regulation 6). The author of a database may therefore enjoy copyright in it and may be able to sue third parties for infringement. By virtue of Regulation 13(1) the maker of a database (whether or not it is a copyright work) acquires a property right called a “database right”, if he has made a substantial investment in obtaining, verifying, or presenting its contents. By Regulation 16 a person infringes that database right if, without the consent of the owner, he extracts or re-utilises all or a substantial part of the contents of the database. The sections of the 1988 Act relating to dealing with rights in copyright works and the rights and remedies of copyright owners apply in relation to a database right and the database in which that right subsists (Regulation 23). The nature of the protection accorded to the makers of databases by the 1988 Act and the Regulations reflects a clear recognition that databases do not represent tangible property of a kind that is capable of forming the subject matter of the torts that are concerned with an interference with possession.

Mr. Cogley put forward four separate arguments in opposition to that analysis. The first was that the database in the present case should be regarded as a physical object, because it exists in a physical form on the data manager’s servers. The second was that the essence of possession is physical control coupled with an intention to exclude others and that a person can properly be said to possess something if he is able to exercise complete control over access to it. An example might be the possession of goods in a warehouse to which there is only one key. The third was that a database can be regarded as a document and can be treated as if it were one for all purposes. The fourth was that there is a distinction to be drawn between choses in action, properly so called, and other kinds of intangible property, such as an electronic database, to which the principles enunciated on OBG v Allan do not apply.

As to the first of these submissions, I fully accept that entering information into an electronic data storage system results in an alteration to the physical characteristics of the equipment. It is unnecessary to discuss the details of the processes by which information is stored in, and retrieved from, computers. It is sufficient for present purposes to say that in one way or another (depending on the storage medium) physical changes are brought about in the storage medium which embody the entry of the information and enable it to be recalled. In that sense the process is similar to making a manuscript entry in a ledger: there is a physical change in the condition of the ledger by the application of ink to a sheet of paper. However, that does not in my view render the information itself a physical object capable of possession independently of the medium in which it is held and in the electronic world the distinction is of some importance because of the ease of making and transmitting intangible copies.

The distinction between a disk or other medium on which data is held (which is a tangible object) and the data itself (which is not) was clearly recognised by this court in St Alban’s City & District Council v International Computers Ltd [1996] 4 All E. R. 481. It was followed and applied by Patten J. (as he then was) in Thunder Air Ltd v Hilmarsson [2008] EWHC 355 (Ch) (unreported), in which an application was made under section 4 of the Torts (Interference with Goods) Act 1977 for an interim order for the delivery up of various materials, including documents held in electronic form, in connection with a claim for conversion and an order for delivery up under section 3. The relevant documents were alleged to be in the possession or control of the defendant, but it was accepted that documents stored in electronic form do not constitute goods within the meaning of the legislation. The judge held (paragraph 29) that there could therefore be no claim under the Act for wrongful interference with (and thus for delivery up of) any documents stored on the defendant’s computers.

The contract in the present case is silent on the mechanics of the transaction. As a result, it is not clear exactly what was to be transferred by the publisher to the data manager, or how, or what was to happen when the contract came to an end. However, the appeal was argued on the basis that the parties intended the relevant data to be transferred electronically by the publisher to the data manager (i.e. without the use of physical media such as disks or other forms of portable storage devices) and held by the data manager on its own computers or computers to which it had unrestricted access and I am content to assume that that was so. No one suggested, therefore that there was to be a physical transfer of equipment of any kind at any stage. In my view, therefore, the fact that the transfer of the data to the data manager resulted in a physical alteration to its own systems or to systems to which it had access takes the case nowhere.

The second of Mr. Cogley’s submissions has greater attraction. The range of circumstances in which a lien has been held to arise is broad, encompassing many different circumstances in which work has been carried out on, or in the production of, property of one kind or another: see the paragraph in Halsbury to which I referred earlier. I am willing to assume for present purposes that, if the database in this case were tangible property, the data manager would in principle be entitled to exercise a lien over it for outstanding fees. Mr. Cogley submitted that in those circumstances it is irrelevant that the database is incapable of physical possession; what matters is whether the data manager is able to exercise effective control over it as against the publisher.

Although an analogy can be drawn between control of a database and possession of a chattel, I am unable to accept Mr. Cogley’s argument. It is true that practical control goes hand in hand with possession, but in my view the two are not the same. Possession is concerned with the physical control of tangible objects; practical control is a broader concept, capable of extending to intangible assets and to things which the law would not regard as property at all. The case of goods stored in a warehouse, the only key to which is held by the bailee, does not in my view undermine that distinction, because the holder of the key has physical control over physical objects. In the present case the data manager was entitled, subject to the terms of the contract, to exercise practical control over the information constituting the database, but it could not exercise physical control over that information, which was intangible in nature. For the same reason the withholding of the database by the data manager could not, even if wrongful, constitute the tort of conversion.

Next Mr. Cogley submitted that a database can and should be treated for all purposes as a document, drawing the court’s attention to the decision of Vinelott J. in Derby & Co. Ltd v Weldon (No. 9) [1991] 1 W.L.R. 652. In my view this argument proceeds on a false basis. The former Rules of the Supreme Court, and now the Civil Procedure Rules, provide for the disclosure in proceedings of documents in the possession, custody or power of the parties. The rules originated long before the creation and storage of documents in electronic form had been conceived of and, not surprisingly, once electronic documents came into existence steps were taken to ensure that documents stored on electronic media were subject to the same rules. It was unnecessary for that purpose for the rule makers or the courts to concern themselves with the precise nature of the information or the manner in which it was held. They were not concerned with the niceties of possession but with the ability of parties to provide, usually, but not necessarily, in paper form, copies of documents held in electronic media. There was no difficulty in understanding what “possession, custody or power” meant in that context, but it is a far cry from that to hold that documents held in electronic form are capable of being possessed in the sense in which that the word applies to chattels. In my view no assistance is to be gained from this line of argument.

Mr. Cogley’s fourth argument was that, if the database cannot be regarded as a physical object for these purposes, it is a form of intangible property different from a chose in action. As such, it is capable of being possessed and wrongful interference with it will constitute the tort of conversion. I am unable to accept that. In Colonial Bank v Whinney (1885) 30 Ch. D. 261, the court had to decide whether shares in a joint stock company were to be classified as choses in action for the purposes of the proviso to section 44(iii) of the Bankruptcy Act 1883, by which property in the order or disposition of the bankrupt in his trade or business with the consent of the true owner, other than choses in action, was made available for the satisfaction of his debts. In the Court of Appeal the case provoked a great deal of learned debate about the history and nature of choses in action, much of it contrasting choses in action with choses in possession. In the end Cotton and Lindley L.JJ. held that shares were not choses in action for the purposes of the statute, although they both regarded them as a form of intangible personal property. Fry L.J. dissented. In his view “all personal things are either in possession or in action. The law knows no tertium quid between the two.” He therefore held that shares were choses in action.

The view of Fry L.J. was subsequently preferred by the House of Lords (1886) 11 App. Cas. 426. Lord Blackburn, who gave the principal speech, noted that there had always been a difference between personal property, which was capable of being stolen, taken, and carried away, and thus the subject of larceny at common law, and other kinds of personal property which could not be the subject of larceny or be taken in execution, because they could not be seized. In my view that decision makes it very difficult to accept that the common law recognises the existence of intangible property other than choses in action (apart from patents, which are subject to statutory classification), but even if it does, the decision in OBG v Allan prevents us from holding that property of that kind is susceptible of possession so that wrongful interference can constitute the tort of conversion. It follows, in my view, that it is equally not susceptible to the exercise of a possessory lien.

Many of Mr. Cogley’s arguments owed a debt to a scholarly volume entitled The Tort of Conversion, in which Sarah Green and John Randall Q.C. make a powerful case for recognising that the essential elements of possession can be exercised over digitised materials, of which a database is a prime example, which should therefore be amenable to the tort of conversion. They also make a powerful case for reconsidering the dichotomy between choses in possession and choses in action and recognising a third category of intangible property, which may also be susceptible of possession and therefore amenable to the tort of conversion. Inevitably they are critical of the decision in OBG v Allan, which they regard as a wasted opportunity to set the law on a modern footing. In my view there is much force in their analysis, which, if accepted, would have the beneficial effect of extending the protection of property rights in a way that would take account of recent technological developments. However, to take the course which they propose would involve a significant departure from the existing law in a way that is inconsistent with the decision in OBG v Allan. That course is not open to us – indeed, it may now have to await the intervention of Parliament – and I do not think that any purpose would therefore be served by embarking on a fuller discussion of their suggestions here.

Mr. Susman submitted that there were two other reasons why the data manager in this case could not exercise a lien over the database. The first was that the work it carried out in this case was more in the nature of maintenance than improvement of the database and that for that reason alone it was not capable of supporting a lien: see Hatton v Car Maintenance Co Ltd [1915] 1 Ch. 621. The database consists of many individual pieces of information stored in a structured manner which renders them capable of systematic search and retrieval. The data manager’s task was to make whatever amendments were necessary from time to time to ensure that the information it contained was (so far as reasonably practicable) correct at all times. The line between maintenance and improvement can be a fine one. If the database had not been updated regularly, it would soon have lost much of its value. In that sense it might be described as a wasting asset as the information it contained became more and more out of date. On the other hand, the updating necessarily involved the entry of new information and to that extent it can be said that it involved improvement. In the end I do not think that this question is determinative of the appeal, but if it were, I would be inclined to hold that the updating of the database was more in the nature of improvement than mere maintenance.

The second reason was that it would be contrary to the contract for the data manager to exercise a lien. It was common ground that if, on its true construction, the contract under which the property is delivered to the bailee is inconsistent with the exercise of a lien, no such right will arise. One difficulty in the present case, however, is that the contract says nothing about the publisher’s right of access to the database during the currency of the contract or how the database will be dealt with when the contract comes to an end. Nor, unfortunately, were findings made by the judge below which provide much assistance towards resolving this aspect of its construction.

However, it seems to have been accepted by both parties that the publisher was entitled to have access to the database at will during the currency of the contract and the data manager apparently provided it with a password for that purpose. That may have reflected a recognition on both sides that access was necessary for the purposes of the publisher’s business, in which case it is likely that a term giving the publisher a right to unrestricted access might easily be implied. The contract is so brief and informal that a full understanding of the background would almost certainly lead to the implication of a number of terms, simply to give effect to what the parties must be taken to have intended, but we do not have the benefit of the kind of findings that would enable us to reach a conclusion on such matters with any degree of confidence. All we know is that the data manager did not in fact exercise exclusive control over the database during the currency of the contract, although, as events later demonstrated, it could effectively have done so by changing the password and withholding the new password from the publisher.

Before he can exercise a lien at common law a bailee must have obtained a continuing right of possession which he is entitled to exercise against the bailor. Thus a racehorse trainer cannot exercise a lien over a racehorse for his fees if the contract reserves to the owner (expressly or by implication) the right to decide the places at which and the jockeys by whom it is to be raced: see Forth v Simpson (1849) 13 Q.B. 680. Likewise, one reason given for denying to a keeper of livery stables the right to exercise a lien for his charges is that he is obliged to give possession of the horse to the bailor whenever requested: see Scarfe v Morgan (1838) 4 M. & W. 270. (Another is that feeding and stabling does not improve the horse: see Judson v. Etheridge (1833) 1 Cromp. & M. 743 and In re Southern Livestock Producers Ltd. [1964] 1 W.L.R. 24.) Although the contract in the present case contained no express provision for the publisher to have access to the data, neither did it contain any provision, express or implied, excluding him from it and the fact that the data manager did in fact make access to it freely available by the provision of a password is in my view inconsistent with the conclusion that he was in fact exercising the kind of exclusive control that would equate to the continuing possession required for the exercise of a lien. In view of the other conclusions to which I have come it is not necessary to reach a final decision on this point, but if necessary I would hold that in this case the data manager did not exercise the degree of control necessary to entitle it to exercise a lien.

In those circumstances it is unnecessary to discuss at any length the question whether, if it were open to us to do so, it would be desirable extend the reach of the common law lien to electronic data. In Scarfe v Morgan Parke B. expressed the view that particular liens “being consistent with the principles of natural equity, are favoured by the law, which is construed liberally in such cases.” That may be so, but I cannot see any basis on which the extension of the right to exercise a lien over intangible property could rationally be confined to electronic databases and for my own part I am not persuaded that it is necessary or desirable to extend this form of self-help, based on control rather than possession, to intangible property generally. The majority view in OBG v Allan suggests the contrary. Transfers of intangible property, whether in electronic or other forms, will almost invariably be covered by contracts which, if the parties so wish, may provide expressly for situations of the kind that arose in this case.

For the reasons I have given I would hold that the data manager was not entitled to exercise a common law lien on the database. It may well be that a fuller understanding of the background to the contract would support the conclusion that the parties intended the publisher to have access to the database at will, but whether that is so or not, it must have been implicit in the contract that when it came to an end the data manager was under an obligation to send the publisher by electronic means a copy of the database in its latest form. (We are not concerned with any term that might be implied in relation to the removal of the copy of the database remaining in the data manager’s own systems.) It was therefore in breach of contract in refusing to do so, unless the contract impliedly gave it a right to withhold that information from the publisher and to exclude the publisher from its systems until it had received payment of any outstanding fees. For my part I do not think that there is a sufficient basis for implying a term of that kind.

Although a right to withhold access to the database would not give the data manager any proprietary interest in it, it could in the case of an insolvency effectively enable him to assert a preferential position to the disadvantage of other creditors. Whether an extension of the tort of conversion to cover interference with intangible property is desirable, I am not persuaded that it is desirable to extend the scope of the common law lien in the way suggested in this case. Although in former times the law may have looked with favour on possessory liens as reflecting a form of natural equity, the parties to a contract of the kind under consideration in this case are free to make express provision for their rights and obligations on termination of their relationship and an extension of the right of self-help is to that extent less justifiable. In any event, for the reasons I have given I do not think it is open to us to take that course, even if we wished to.

The contract made no provision for the manner of its termination, but the parties accepted that by implication reasonable notice was required on either side and the judge so held. The only remaining question was what constituted a reasonable period for those purposes. The judge held, correctly, that that was a question of fact, not a question of law, and therefore he was entitled to consider any evidence that could properly be regarded as having a bearing on it. In reaching his decision the judge relied quite heavily on an assertion made by Mr. Kayani, the publisher’s managing director, during a telephone conversation with Mr. Hooker in May 2011 that three months’ notice would be reasonable. In my view he was entitled to do so, despite the fact that the conversation took place some time after the contract had been made. At that stage the judge’s task was not to construe the contract; it was to decide as a matter of fact what was a reasonable period of notice.

For the reasons I have given I would allow the publisher’s appeal to the extent of holding that the data manager was not entitled to refuse to provide the publisher on request with a copy of the database in its current form and was in breach of contract in doing so. I would dismiss its appeal in relation to the period of notice.

I also would allow this appeal to the extent indicated, for the reasons given by Moore-Bick L.J., with whose judgment I wholly agree.

The sub-text of Mr. Cogley’s submissions was that the courts should not leave the common law possessory lien stuck in its eighteenth and nineteenth century origins and development; rather the courts should go on to give it a twenty-first century application, appropriate to modern times and modern commercial activities. That appeal to modernism has its attractions: indeed, it was that approach which seems to have decided matters on this issue so far as the district judge was concerned.

But I think the court should resist those attractions. I say that for two connected reasons.

The present case, at all events, underlines the desirability of parties to business dealings such as these including express contractual provisions to cover the position in the event of termination.

I agree with Moore-Bick L.J. that, for the reasons he has given, the publisher’s appeal must be allowed to the extent he has indicated. I also agree with Davis L.J. that the potential unintended consequences constitute a further reason for not taking the step which we were invited to take by the respondent.

I would add only one observation in connection with the wider implications of Mr. Cogley’s submission that the electronic database was a type of intangible property which, unlike choses in action, was capable of possession and thus of being subject to a lien. An electronic database consists of structured information. Although information may give rise to intellectual property rights, such as database right and copyright, the law has been reluctant to treat information itself as property. When information is created and recorded there are sharp distinctions between the information itself, the physical medium on which the information is recorded and the rights to which the information gives rise. Whilst the physical medium and the rights are treated as property, the information itself has never been. As to this, see most recently per Lord Walker in OBG Ltd v Allan [2007] UKHL 21, [2008] 1 A.C. 1 at [275], where he is dealing with the appeal in Douglas v Hello, and the discussion of this topic in Green & Randall, The Tort of Conversion at pages 141-144. If Mr. Cogley were right that the database could be possessed and could be the subject of a lien and that its possession could be withheld until payment and released or transferred upon payment, one would be coming close to treating information as property. That observation further underlines the significance of the step we were invited to take.