S.4 of the Statute of Frauds is no “dusty relic”. This much was conceded by the unsuccessful appellants in Golden Ocean Group Ltd v Salgaocar Mining Industries PVT Ltd [2012] EWCA Civ 265, a case dealing with the enforceability of a guarantee given and proved by a chain of e-mail correspondence. The language of the Statute may be arcane, but as Lord Hoffmann pointed out in Actionstength Ltd v International Glass Engineering SPA [2003] 2 AC 541, the policy of protecting people from being held liable as guarantors on the basis of ill-considered, ambiguous or completely fi ctitious oral utterances remains good. Translated into modern terminology, s.4 requires that, for a guarantee obligation to be enforceable: (1) the agreement to guarantee must be in writing or, if the agreement is made orally, there must be a memorandum or note evidencing the oral agreement, and (2) the agreement, memorandum or note must be signed by the guarantor or someone authorised by him to sign it on his behalf. In an electronic age, the courts must grapple with the thorny issues that are raised by allegations that guarantee obligations have been created by instantaneous electronic communication. The FAQs that follow are intended both as a guide and as a warning.

  1. Can a guarantee be concluded, or evidenced, by e-mail correspondence?

The practical problem with e-mail is also its greatest benefi t. It is instantaneous. It can feel like a conversation, in which touchtyping replaces speech. But an e-mail is clearly a piece of writing. Although the draftsman of the Statute of Frauds could not have foreseen modern developments in information technology, guarantees had in the Twentieth Century been enforceable if made by fax. Article 9(1) of EC Directive 2000/31 requires EU Member States to ensure that their legal systems permit contracts to be concluded electronically, and the Law Commission expressed the view in 2001 that in general the common law was suffi ciently fl exible to ensure compliance with this obligation. It took Judge Pelling QC (sitting as Judge of the Chancery Division) in J Pereira Fernandes SA v Mehta [2006] 2 All ER 891 to confi rm the obvious proposition that an enforceable guarantee obligation may be created or evidenced electronically.

In Mehta, the allegation was not that there was an e-guarantee in writing, but that the e-mail correspondence showed a suffi cient memorandum or note. Nilesh Mehta had authorised a member of his staff to send an e-mail to the creditor at a time when a winding-up petition had been presented against a company of which Mr Mehta was director. By the e-mail, Mr Mehta sought an agreement for an adjournment of the hearing of the petition, and offered that he would give ”A Personal Guarantee ... in the amount of £25,000” to the creditor. The e-mail concluded “I am also prepared to give a company undertaking ... pending the signing of the Personal Guarantee”. The e-mail was an offer to give a guarantee, which was then orally accepted by the creditor. By analogy with Parker v Clark [1960] 1 WLR 28, the written (e-mailed) offer was held to be a suffi cient memorandum for the purpose of s.4. If an e-mailed memorandum suffi ces for s.4, it follows that an e-mail is capable also of comprising a written guarantee: this is confi rmed by Golden Ocean.

Parliament could legitimately have chosen not to allow certain kinds of guarantees to be created, or evidenced, by e-mail. Article 9(2) of EC Directive 2000/31 allows Member States to provide special rules for “contracts of suretyship granted ... by persons outside their trade business or profession”. Parliament did not respond by exempting “consumer” guarantees. Instead, the Consumer Credit (Electronic Communications) Order 2004 expressly allows guarantees regulated by the Consumer Credit Act 1974 to be concluded electronically, and makes provision for the form and content of such regulated security documents.

  1. Can an e-guarantee be concluded or proved by consideration of a long thread of emails?

In Golden Ocean, Christopher Clarke J at fi rst instance had held that an enforceable contract of guarantee of obligations under a charterparty may be created by an electronic chain of documents. The Defendant guarantors had applied to set aside the Claimant creditor’s permission to issue a Claim Form and serve it on them in Goa, on the basis among other things that the e-mail correspondence on which the creditor relied as creating the guarantee obligation was not contained in a single electronic document. To fi nd a guarantee, one had to work backwards from an e-mail of acceptance reading “Yes. Confirm the 5 days that’s fi ne. Cd U send me recap – with todays date?”, through a series of other e-mails, eventually to reach an e-mail sent some weeks earlier which described the recap as “A/c Trustworth Limited Singapore fully guaranteed by Salgaocar Mining Industries Goa”.

On appeal, it was argued that it would be inimical to the purpose of the Statute of Frauds that it be satisfi ed only after a lengthy, educated trawl through a lever-arch fi le full of email exchanges. One might comment that the volume of documentation might in most modern commercial cases be reduced if parties to litigation found a way round the habit of printing out every single chain behind every single e-mail each time that each e-mail produced a reply. But leaving that aside, and notwithstanding that the submission was just forensic exaggeration (the e-mail chain was relatively short), the Court of Appeal held that the length of the chain was not important. Tomlinson LJ said at [22] that “I can see no objection in principle to reference to a sequence of negotiating emails or other documents of the sort which is commonplace in ship chartering and ship sale and purchase”. Tomlinson LJ approved the first-instance decision that there is no principled limit to the volume of documentation through which one might have to trawl in order to determine whether a contract of guarantee had been concluded. There would not appear to be any rationale for confi ning the decision to the immediate context of chartering and maritime contracts; it may be the case that the chartering context gave precise meaning to terms employed by the negotiators and made it unambiguous that the parties intended to be bound by their relatively informal e-mail correspondence, but the principle is one of general application.

  1. Is an electronic signature a “signature”?

Unsurprisingly, yes. In Golden Ocean, it was common ground that an electronic signature was good enough. In Re Stealth Construction Ltd [2012] 1 BCLC 297, it was conceded by the applicant liquidators that the insertion of the correspondents’ fi rst names at the bottom of the e-mails in question was suffi cient for the purpose of s.2 of the Law of Property (Miscellaneous Provisions) Act 1989, which also imposes a signature obligation in order that an enforceable agreement be created. The concessions are clearly correct.

  1. What is a sufficient “electronic signature”?

S.7(1) of the Electronic Communications Act 2000 provides for the admissibility into evidence of any “electronic signature” which has been incorporated into or is logically associated with a particular communication and which has been “certifi ed” by the signatory. S.7(2) goes on to defi ne “electronic signature” in somewhat circular fashion as something in electronic form which (i) is incorporated into or logically associated with an electronic communication or data, and (ii) purports to be so incorporated or associated for the purpose of being used in establishing the authenticity of the communication or data and/ or its integrity. On the face of it, this is much too circular to be of any practical use, and concerns anyway the law of evidence and not the Statute of Frauds.

Flaux J held in Lindsay v O’Loughnane [2010] EWHC 529 (QB) at [95] that s.6 of the Statute of Frauds (Amendment) Act 1828 (under which a fraudulent misrepresentation as to credit is not actionable unless the representation is made in signed writing) would “clearly be satisfi ed” provided that the representation was contained in an e-mail which “includes a written indication of who is sending the e-mail”. That last phrase is perhaps a little wide. The cases suggest that a mere “written indication” is not enough to constitute a signature for statutory purposes. To satisfy s.4 of the Statute of Frauds, a purported signature must be one which “is intended for a signature”: see Evans v Hoare [1892] 1 QB 593. In Decouvreur v Jordan (Times, May 25, 1987), the Court of Appeal held that “any writing by the party to be charged by which he identifi es himself or by which he can be identified ... and which shows, objectively, an intention to adopt the note or memorandum will suffice”.

So, the question is whether a given “written indication of who is sending the e-mail” is, objectively, one which demonstrates an intention to have the sender identifi ed as adopting the document. This concurs with the view of the Law Commission that compliance with any statutory signature requirement for an electronic document can be tested in a functional way by analysing whether the conduct of the would-be signatory manifests an intention to authenticate the relevant instrument.

The recent cases demonstrate that a suffi cient electronic signature may be given very informally. In Re Stealth, “Jo and Suzy” were enough. In Golden Ocean, the e-mail at issue, which had purportedly been sent on behalf of the Defendant company, also contained the sender’s fi rst name (“Guy”) at its foot. The e-mail’s language was described as “matey”, but the communications were not merely inconsequential. Tomlinson LJ insisted that the decision to sign off the acceptance e-mail with the word “Guy” was properly regarded as an authentication of the contract which was contained in the chain of e-mails of which it was the culmination. Indeed, even a nickname may be enough, so long as it is a suffi cient identifi cation with a suffi cient objective intention to authenticate.

By contrast, in Mehta, Judge Pelling did not accept that an e-mail memorandum which did not contain Mr Mehta’s name anywhere than in the “sent by” box appearing in the recipient’s inbox had a suffi cient signature. The Judge relied on Lord Westbury’s speech in Caton v Caton (1867) LR 2 HL 127, which differentiated between a “signature” appearing in an instrument only incidentally, and a “true” signature which is intended to relate and refer to every part of the instrument in question. Judge Pelling considered at [29] that, “absent evidence to the contrary” the automatic insertion of an e-mail address could not be held to have been “intended for a signature”. The decision is obviously right, but the proviso gives rise to some problems: what sort of evidence could Judge Pelling have had in mind, other than that of the subjective intention or purpose of the sender? The solution may be that Judge Pelling was cautiously allowing for circumstances that may develop in which a party is able consciously to choose whether to allow his e-mail address to be read by the recipient; where such an election is available and known to be available, then it may be said that an objective intention to sign may be manifest.

This chimes with Christopher Clarke J’s explanation of Caton in Golden Ocean at fi rst instance ([2011] 2 All ER (Comm) 95) at [95]: “there must be something ... which is voluntarily affi xed to the document by way of authentication thereof” (emphasis added). A formulation which includes a proviso of voluntariness does throw up some problems of its own where automatically-generated e-mail footers are involved. The lowly employee in a large corporation may in practical terms have no choice about the form of the automatic footer, and no realistic method of having it removed. Can he be said to have voluntarily affi xed the footer to the document? To the extent that the problem will arise in a personal guarantee case, the answer is probably yes. It appears that in Lindsay v O’Loughnane, just such an automaticallygenerated signature was held to be suffi cient, although the judgment is not wholly clear on the point. From the point of view of the recipient, there will be no difference between (i) the sender who consciously chose to include the signature or not to have it removed, (ii) the sender who did not know that the signature would appear, and (iii) the sender who did not want the signature to appear but was not in a position to remove it (assuming he took no steps to draw this to the attention of the recipient). The sender in all the above examples will objectively have shown an intention to adopt the writing by an automated signature; it is not for the recipient to inquire further.

The unwilling purported e-guarantor who is unable to remove the signature would be well-advised to add rubric at the end of his e-mail to the effect that the automatic signature is not intended to authenticate anything in the body of the text, and hope that has the desired effect. Those with the fi nancial or technological capacity might usefully stipulate from the outset that no message is to be treated as authenticated without an encrypted signature.

  1. Can a guarantee be sent by text? Or Facebook? Or Twitter?

Nowadays, people like to spend most of their time avoiding the exigencies of speaking to one another by communicating in an electronic text-speak which contains only the vestiges of written language. Others brief “friends”, real or otherwise, over remote social networks. On one view, these forms of communication are far closer to oral exchanges than to written exchanges. But the same might once have been said of e-mail. For the purposes of the Statute of Frauds, it must be determined whether or not a given communication is in writing. There is no hybrid category.

Electronic writing is writing for the purpose of the Statute of Frauds, and it would be bizarre if the fact that the writing appears on a particular platform is a suffi cient distinguishing feature. I suspect that the issue will not be one of whether there is “writing”, but of whether there is an intention to create legal relations. At least as they are presently used, Facebook and Twitter (for instance) are not platforms from which one carries out commercial transactions. But the courts’ attitude will change as technologies develop and usage diversifi es. The prevalence of text-message evidence in civil cases in my current practice shows that the courts are treating texts with more and more (commercial) seriousness. “GTEE INVCE PD” texted by a buyer referring to his fi nancially-distressed principal’s purchase might be enough (although “GTEE INVOICE PD LOL” would probably not get home). The Twitter-user who tells his umpteen followers that he has ”just guaranteed £million supply of widgets by A to B” may fi nd himself having created a written memorandum: there is no requirement that the creditor even see the memorandum. In the light of Mehta and Golden Ocean, the only sure-fi re answer – for a purported e-guarantor – is the oldest one in the (physical, metaphorical, but defi nitely not e-book): never sign anything.