For the last ten years Richard Stephens has provided an annual update on contract law for the SCL and also offers it by way of internal training for law firms and in-house legal departments. That puts him in a superb position to take a look at what has happened over the last decade and think about the major themes in contracts and contract law over those years. This is the latest in our series of articles from the Tech Law Masterclass appearing in the October/November issue of Computers & Law.
I shudder to think how many cases I have read - the worst are the long cases I have read coming out of the TCC which I dutifully read before deciding that there is nothing worth reporting on. The last update I did for the SCL back in February 2018 looked at dozens of cases in itself and over a decade I have lectured on hundreds of them, all of which I chose because they had the potential to change drafting, practice or handling disputes in some way.
What is striking about looking at so many cases and talking about them? I suppose a few features come to mind.
Why is it that cases looking at a similar point in the law come along all at once? Take the example of Virulite v Virulite  EWHC 366 (QB) which was just about the first recorded case on the question of what have become known as No Oral Modification clauses. The judge in Virulite decided that such clauses were in fact ineffective to prevent subsequent oral modifications. Suddenly, we had Globe Motors v TRW Lucas  EWCA Civ 396, where the Court of Appeal thought (obiter) the same thing, and MWB Business Exchange Centres v Rock Advertising  EWCA Civ 553, where the Court of Appeal actually concluded likewise. It then went to the Supreme Court, which pulled a rabbit out of the hat and decided that, of course, a No Oral Modification clause was indeed effective to prevent a subsequent oral modification (see  UKSC 24).
Why did these cases come along within just a few years, to decide a point which had never really arisen in previous decisions? This probably leads to the next point …
The law follows our drafting: once something becomes embedded in template commercial contracts (which seem, despite copyright laws, to do the rounds) then you get a series of cases telling us what our drafting really meant all along. That may or may not match the presumptions of the original drafters. I remember years ago that someone put together an article collating all the cases he could find in the law reports on the subject of reasonable (or whatever) endeavours: the result was some half a dozen cases. It seems that there are that number coming out of the courts now every month dealing with the same wording.
The most obvious example of where drafting can create subsequent law is the use of ‘good faith’: this was an expression I almost never saw when I started out in the law, but now it has become ubiquitous in commercial contracts. The first English case in the modern era on this was Gold Group Properties v BDW Trading  EWHC 1632 (TCC), where the contract included (as they do so frequently these days) a clause specifically importing an obligation of good faith. The deputy High Court judge picked his way among Australian authorities and such few authorities as the English courts had by then thrown up and concluded that such an obligation did not supersede specific obligations in the contract (such as the pricing). This was followed in quick succession by numerous cases considering the same expression and it was only a short step from there to start to consider whether or not English Law recognised a pre-existing doctrine of good faith (with some judges, notably Leggatt LJ, very keen on the idea). Is there, isn’t there? What seems clear to me is that the drafting led to the law. Be careful of what you wish for. This leads to the next point …
There are surprises along the way: once the courts get the bit between their teeth, there is no knowing which direction the law will take. Take the example of No Oral Modification clauses I gave above, or the emergence of good faith as a (possible) independent doctrine in English Law. There has emerged in academic circles in recent years a theory of “relational” contract law - the idea that the law is developing in response to ever longer contracts and that separate doctrines of law will need to emerge to service such beasts.
One example of where this theory came into the practical arena was in Amey Birmingham Highways v Birmingham  EWCA Civ 264 concerning a 25 year PFI contract, in which the Court of Appeal had to step through a contract of some 5,190 pages (excluding documents incorporated by reference), of which some 200 pages were devoted to definitions alone. The Court of Appeal had to consider ingenious arguments culled by the contractor from the inevitable infelicities thrown up by such a long contract before deciding that common sense was required in approaching a ‘relational’ contract of this nature and that it was not open to a party to cobble together arguments based on disparate passages in a prolix and multi-authored contract.
Going alongside this development, there have been quite a number of cases where the contract has gone wrong for some reason or another, and the court has to decide what the clause in question really means because the clause is obviously badly drafted. Many of the cases which deal with construction touch on this point, but there have also been some which demonstrate the shortcomings of modern legal practice. Milton Keynes v Viridor  EWHC 239 (TCC) is one such, where the luckless assistant solicitor put an unpopulated document, which had been provided to her by the consultants, into the final agreement, thereby making a nonsense of the charging provisions. The judge described this as “sloppy work” by the consultants who had provided the document in the first place, and to a lesser extent by the solicitor herself. In the modern hothouse that is legal practice, who can be surprised when mistakes of this sort occur? It is surely no coincidence that in the same year, we saw Sutton Housing Partnership v Rydon Maintenance  EWCA Civ 359, the case which was about construing a Service Level Agreement which included worked examples of how the service levels should be applied but omitted any actual service levels.
Is Artificial Intelligence really going to provide a solution to human error - would a computer make the same mistake? The whole point of AI is that it is able to ‘read’ the entirety of our modern prolix documents and make sense of them, without the human eye glazing over as it approaches Schedule 13.
Watch this space, which leads to the next point …
Law never used to be taught as an academic subject until relatively recently - before the Second World War, law was something the novice was expected to pick up by practical experience, in articles or pupillage as the case might be. This would no longer be adequate in today’s environment given the wave after wave of regulations and commentaries on regulations, best practice guidelines and so on (think: GDPR). Just as contracts have got longer, so have cases and legal textbooks. What surprises me as I go through so many cases is how the higher courts have come to be repositories of case-law going back decades or even hundreds of years, back to the early nineteenth century or even earlier.
When Cavendish Square Holding v Makdessi  UKSC 67 got to the Supreme Court, many expected that the judges would dispose of the anomalous rule against penalties as being a rather strange and exceptional intrusion into what is now regarded as an overriding principle of freedom of contract. Not so - the judges went through the legal history books right back to the very beginnings of our legal system to reach their surprising conclusion that there was still a point to keeping the rule against penalties. One of the rules proposed, that we should identify primary and secondary obligations, sounds more like arid text from a dry book on jurisprudence and had the judges themselves in disagreement as to how to apply it. Other judges came up with different rules for identifying a penalty. We might expect the law to be clear, but it is not.
The decision of the Supreme Court in Rock Advertising (see above) turned its back on common law precedent from this jurisdiction and around the common-law world and simply decided on principle that various international conventions did the job rather better (notwithstanding that the UK had not ratified one of them). I wonder whether in 100 years the courts will still keep going back to the contract cases of the nineteenth century or whether they will draw an arbitrary line and simply refuse to consider cases before, say, 1980?
I was always struck by the Roman Law of Citations issued by Emperor Valentinian III in AD 426: basically, Roman Law had got itself stuck in a quagmire of its own making, with too many regulations and juristic opinions, meaning that the law could mean anything in the hands of the right advocate. The Law of Citations made quick work of the problem: reduce the number of permitted juristic opinions to five and then decide the law according to the majority view or, in the case of conflict, to an order of precedence among the jurists (Papinianus ranking as the number one jurist). If we had to choose five commercial judges of the modern era, which five would qualify for a modern English Law of Citations? And who would be the modern Papinianus, ranking above all the rest?
What are the cases which have really taken me by surprise?
Looking back over ten years, where have the big changes been?
As IT lawyers are so fascinated by anything on liabilities, I will lead off with this one. A whole run of cases in the Court of Appeal has decided that there is no special approach to interpreting liabilities provisions, no need in the vast majority of cases for a principle of contra proferentem, and that all you had to do was construe the words on the page. This is a complete reversal of the law I learned at university.
Leading on from that is the clause that most lawyers incorporate into their contracts excluding consequential loss. If you ask a commercial lawyer what that means, some may tell you it excludes the losses in the second limb of Hadley v Baxendale  EWHC Exch J70 (1854) 9 Ex Ch 341 156 ER 145 though fewer will be able to explain the rule and fewer still will be able to give a good illustration of how the ‘rule’ is applied. Along comes the decision in Star Polaris v HHIC-PHIL  EWHC 2941 (Comm) which decides that consequential does not mean that at all, but is simply a term to be interpreted in context. In other words, it has no fixed meaning.
Most templates, however, still exclude ‘consequential loss’.
Entering into a contract
One of the aspects of reviewing the common law over a period of time is how the cases come and go in a sort of ebb and flow: at first, a decision will ‘clarify’ the law, which in turn opens up a cottage industry in seeking to apply that law until the original rule is forgotten and needs to be re-stated by a hard case. Such was the fate of RTS Flexible Systems v Müller  UKSC 14 which sought to lay down the law on the all too common problem of where the parties start work on a project without having a signed contract. A common enough problem and you would expect the law to be easy to state and simple to apply - but not so. In the ‘contract or not’ cases, the answer came almost inevitably to be ‘contract’” rather than ‘not’ until we got the cases of MacInnes v Gross  EWHC 46 (QB) where the claimant alleged a contract made over dinner at Zuma off Knightsbridge and Blue v Ashley  EWHC 1928 (Comm) where the claimant alleged a contract made in the course of a somewhat heavy drinking session at a pub. Both failed - along with some other cases recently - perhaps indicating that the courts have had enough of successful claims for a finding of a contract and the start of the requirement for more formality.
The same point could be made about the cases on ‘correction by construction’. In the modern era we had Investors Compensation Scheme v West Bromwich  UKHL 28, where the modern law was laid down by Lord Hoffman. This led to Chartbrook v Persimmon Homes  UKHL 38, which showed the degree of change a court was prepared to make, effectively re-writing a crucial clause to make it say what one party said it should mean, though it did not say so on its face.
This again led to a cottage industry in claimants saying that contracts did not mean what they apparently said on the face of it: it was hard work for me during those years trying to summarise for the purposes of a lecture what were complex contractual provisions and the elegant, if not Byzantine, arguments deployed to undermine their literal meaning. Then along came Arnold v Britton  UKSC 36, which concluded that contracts did, after all, mean what they said. The waters have been slightly muddied, it is true to say, by the Supreme Court’s consideration of the tortured wording (the Supreme Court used the more diplomatic ‘avoidably opaque’) in Wood v Capita Insurance Services  UKSC 24, where the Supreme Court’s handling of concepts like textualism and contextualism left the field wide open for judges to do more or less what they liked in construing contracts. Perhaps the lesson to learn from this is not to draft clauses like the one in Wood - as I have said before, if you can’t read a clause out in one breath, the clause is probably too long. Modern software which reads out text is a real boon to the drafter and should identify where a clause has become ‘avoidably opaque’.
The power of the precedent
I suppose that is the lesson I draw from a few cases which have looked at order of precedence clauses. Such a clause is really an admission from lawyers that they have lost control of the drafting process and do not really know what inconsistencies and infelicities there are left in the document. After all, they probably only drafted the frontend and left the schedules to the commercial or technical people and frankly do not trust them. Hence an order of precedence which typically gives precedence to the frontend over everything else.
RWE Npower Renewables v J N Bentley  EWCA Civ 150 was the first case to look at this vexed problem but decided (with the Court of Appeal agreeing with the trial judge) that it was not necessary to have recourse to the order of precedence if sense could be made of two apparently conflicting statements in different parts of the overall contract. I was much struck by Alexander v West Bromwich  EWCA Civ 496, which decided to apply literally the injunction in the terms and conditions to give precedence to a mortgage offer letter over the terms and conditions: in this instance, the Court of Appeal held that an inconsistency could arise not just with a clash of words but where the sense and purpose of one document in a contract were inconsistent with another document. By making the mortgage offer letter prevail over the terms and conditions, the Court of Appeal probably came to a decision diametrically opposed to the result intended by the drafters of the terms and conditions.
Finally, in MT Højgaard v E.ON Climate & Renewables UK Robin Rigg East  UKSC 59, the Supreme Court also had to look at an order of precedence clause but in this case construing the obligation in question as meaning that the engineering firm, which followed the customer’s directions in using a flawed international standard, was still responsible for the outcome of the design mandated by the customer. The engineering firm was bound by a warranty covering the lifetime of the design, even though the warranty was buried deep in a technical schedule, and was probably written by an engineer who had no concept of legal warranties or that this sentence would turn out to be the winning clause in a case that went to the Supreme Court. I suspect it was the sort of schedule that never received legal review or advice - but again, the order of precedence provision gave no help to the hapless engineering firm in this case.
However, based on what I see going across my desk, we still use order of precedence clauses.
When I provide in-house training, and I recite the facts of RTS Flexible Systems and say that it took the lawyers in that case several months to settle the last clause, the force majeure provision, it always gets a laugh. However, my own experience is that it is getting harder to negotiate contracts rather than easier. One example of this is the humble notices provision tucked away in the boilerplate. For some reason, lawyers (whose own website describes them as ‘very commercial’ in their approach) seem determined to die in a ditch to preserve the purity of the drafting of their template contract on the question of giving notices.
When I look at the many cases that have come out on this subject in the last few years, I am not sure why. The one lesson they give is that these humble clauses are there to trip us up at a moment of crisis - when there is a dispute and you absolutely have to get a formal notice served correctly, a complex notices provision mandating this or that type of service is a hostage to fortune.
In Teoco UK v Aircom Jersey  EWCA Civ 23 the claimant failed because, after taking two shots at giving notice of a potential claim for breach of warranty following the acquisition of a company, the Court of Appeal agreed with the trial judge in finding that it had not provided ‘reasonable details of the claim’. The claimants failed at first instance in Zayo Group International v Ainger and others  EWHC 2542 (Comm) because they failed to serve all potential defendants with notice of a claim in time, when the notices clause required all defendants to be served and invalidated service which was not effected on all the defendants.
I suppose what comes out of this is that it is impossible to predict the future in detail, only that the law comes and goes in waves - a big case settling the law, then a whole series of cases tests that rule to destruction and then another big case comes along re-stating the law along more cautious lines.
What does this mean for the drafter? I think there is huge value in keeping up to date with the law. Modern legal practice, with its hothouse atmosphere, is struggling to keep up with developments - real changes - which should mandate changes in drafting or, more importantly, proper process. Take First Tower Trustees v CDS Superstores International  EWCA Civ 1396, where the Court of Appeal has said that any non-reliance clause is subject to the reasonableness clause in UCTA: this is not just some tweak to the wording, but a challenge to the whole process of tendering and pre-contractual communication and the lawyer should be advising on the process, not just applying some clever wording to try to cover up errors made along the way.
Again, I suppose I would say that providing this training stops people lapsing into reliance on whatever template they use and start thinking about what they do. Why have we come to use ‘reasonable endeavours’ and ‘good faith’ so frequently without thinking about what they could mean? This applies especially to the latter expression which is seeing significant judicial development and may radically change practice going forward. Why do so many lawyers exclude consequential loss without really knowing what it means or what sort of loss they could be excluding? Why have indemnities become so ubiquitous when most lawyers cannot explain the meaning of expressions such as ‘save and hold harmless’?
In conclusion, we cannot predict what the future holds, but the lawyer should be aware of the current issues and be prepared to advise not just on wording but on process and thereby minimise risk. Oliver Wendell Holmes Jr speaking to an audience back in 1897 said, ‘the prophecies of what the courts will do in fact, and nothing more pretentious, are what I mean by the law’. If we are indeed called upon to make predictions, like weather forecasters, then it behoves us to keep an eye on the prevailing wind and cloud formations.
Richard Stephens FBCS FCIArb, Principal, The Law Office of Richard Stephens – LORS. Richard is an SCL Fellow.
 I say the first - other than a very few oblique mentions in other cases which were not decisive on the point.
 See the UNIDROIT Principles of International Commercial Contracts, 4th ed (2016) and the Vienna Convention on Contracts for the International Sale of Goods (1980) (never in fact ratified by the UK).
 Take a look by way of example at Transocean Drilling v Providence Resources  EWCA Civ 372, Persimmon Homes v Ove Arup  EWCA Civ 373 or Scottish Power v BP Exploration  EWCA Civ 1043.
 At least this gave me a chance to go there to sample its wares and take a selfie for the purpose of giving my annual lecture!
 There are some signs of that with Rosalina Investments v New Balance Athletic Shoes  EWHC 1014 (QB) and Goodwood Investments Holdings v Thyssenkrupp Industrial Solutions  EWHC 1056 (Comm): in most years I could happily spend half a day talking about the ‘contract or not’ cases and listing the cases that had decided that there was indeed a contract notwithstanding the absence of a signed document.