Out-Law Guide | 13 Nov 2007 | 4:14 pm | 6 min. read
Harbinger UK Limited v GE Information Services Limited
The appellant, GEIS, operated networks for customers providing business communications services with an element of security. To do this, it obtained software packages from a variety of sources, including the respondent, Harbinger.
The relevant dispute arose from an agreement dated 19 August 1996. Assignments had taken place since the original agreement but no point was taken on that. The basic arrangement set up by this agreement was common enough: GEIS was a VAR, supplying the product to customers to whom it was authorised to grant sub-licences. A model form of “end-user agreement” was included.
On 30 December 1998, Harbinger gave notice to terminate the agreement with effect from 31 December 1999, and proceedings were commenced shortly thereafter. The report of the decision at first instance is given at  Masons CLR 335, and it will be recalled that HH Judge Thornton determined that this notice was valid in itself. The issue in the appeal related to Harbinger’s liability for continuing support and maintenance.
The full terms of the relevant clause are as follows:
Clause 10 – Support and Maintenance
[Harbinger] will provide support and maintenance in perpetuity for an annual payment as specified in Schedule One, and subject to the following conditions: [there are then some conditions]”
The whole issue in the case revolved around the meaning of what was meant by “in perpetuity”, in particular, in circumstances where the agreement had been validly terminated. Harbinger, foreseeably, argued on the appeal that the obligation to provide continuing support and maintenance terminated with the valid termination of the agreement.
The correct approach to the construction of commercial documents has been re-stated clearly in the decision of the House of Lords in Investors Compensation Scheme Limited v West Bromwich Building Society  1 WLR 896. The judge at first instance was clearly right to look at the overall contractual situation and took into account the model end-user agreement, which itself required GEIS to provide support and maintenance to end users for at least 2 years after delivery. Furthermore, the support and maintenance obligations appeared in the main agreement to be severable from the main obligation to supply the product to GEIS. Thus, “in perpetuity” meant that support and maintenance survived termination by Harbinger of the main agreement.
However, the next question was rather more difficult: for just how long should the support and maintenance obligation then last? At one extreme, GEIS argued that is meant just that – “for ever” or “until the crack of doom”. Put more precisely, the unilateral contract constituted by the agreement lasted for so long as the payments to Harbinger were made by GEIS.
Harbinger’s response was to say that this gave rise to an “uncommercial” result, thus necessitating a search for some shorter period that was still compatible with the words of the agreements. They said that the support and maintenance obligation had therefore to last a “reasonable” time, adopting the interpretation of the judge at first instance.
However, Judge Thornton had not adopted either of the parties’ submissions on this issue. He had observed that GEIS could give notice to its own customers after an initial period of 2 years. Accordingly, he had held that the obligation “in perpetuity” lasted only until the end of that period, i.e. the time when the last of GEIS’ customers’ licences could be validly terminated.
The Court of Appeal disagreed with that conclusion. It was inconsistent with the words “in perpetuity” that there should be some imposed time limit, and the words rather meant “without limit of time”. This did not mean “until the crack of doom”, since technology would become outdated, and customers would require something else anyway. It was just impossible to predict when the last such customer would move to something different. In other words, the full force of “in perpetuity” – that is, without limit of time – had to be given their ordinary meaning, and no restraints of “commercial” reasoning could override this.
Against the argument that it was “uncommercial” for Harbinger to have to provide support and maintenance even for one customer it had to be remembered that Harbinger had been able to advertise the extended support and maintenance and benefit from that.
This is an interesting case, and the first instance decision was reported in an earlier edition of the Masons CLR. There, we commented that Judge Thornton’s solution to the conundrum posed by the drafting of this agreement was a neat and commercial solution.
The Court of Appeal disagreed. The proper construction of agreements has always been a vexed topic, and common law judges have always striven to give sensible effect to what persons in the business community have agreed. The balance of how they do this came up again for consideration in Investors Compensation Scheme Limited v West Bromwich Building Society, where the House of Lords made some interesting observations.
That case arose out of investments that home-owners could make by securing a further advance on a mortgage, and investing the sums so obtained in equities. With rising interest rates and falling equities, this scheme came to grief, leaving many badly affected investors. The claims were taken over by the Investors Compensation Scheme set up by the Financial Services Act 1986, and the case centred on the meaning of clause 3(b) of the assignment of rights of action by an individual investor to the ICS. This clause had an exception, but did it exclude all third party claims or just claims for rescission? The natural meaning of the words excluded all third party claims from the assignment, but this made a nonsense of the agreement. In an earlier case, the judge therefore held that the exclusion was limited to claims for rescission, therefore, against the natural meaning of the words.
Lord Hoffman gave the judgment of the majority of the House of Lords and put forward 5 principles for the construction of documents:
This last point was well made by Lord Diplock in Antaios Compania Naviera SA v Salen Rederierna AB  AC 191, 201:
"if detailed semantic and syntactical analysis of words in a commercial contract is going to lead to a conclusion that flouts business commonsense, it must be made to yield to business commonsense."
In the Investors Compensation Scheme case, for example, the House of Lords looked not only at the (unhelpful) words of the contract in question, but also at the explanatory notes to understand the overall meaning. Similarly, in the Harbinger case, it was not just a question of looking at the words “in perpetuity” in isolation, but rather looking at the whole commercial scheme set up by that agreement and the end user agreements as well, and understanding the legitimate commercial expectations and understandings that the parties would have had.
This makes excellent sense for the parties if courts will take account of these practical matters as well as the actual interpretation of the words. The problem is that two people looking at the same problem will come to different conclusions, as here. Judge Thornton came to a conclusion, based on all relevant factors, and the three judges of the Court of Appeal came to another. Cases abound where matters can go all the way to the House of Lords, with judges divided at all levels, all apparently applying the same canons of construction.
So rules of construction will not lead of themselves to one, and only one, interpretation. Always remember that the commercial purpose must be kept at the forefront of the mind, and the words will be interpreted accordingly. It is refreshing to note that the courts will not give an interpretation at odds with what the parties must have agreed (i.e. wanted to agree) and that one possible option is that court will, in extreme cases, conclude that the parties used the wrong words or syntax. In a sense, the present situation is harder for those called on to construe documents: gone are the days (if they ever existed) when all that was necessary was to look at the words and apply them. Now, any clause in an agreement could be subject to review and analysis in accordance with the parties’ reasonable understanding of their respective positions. As has been seen, rather than easier, this can actually make the process of arriving at the correct interpretation a more difficult process.