Wiser folk than I have often emphasised that in deciding on the correct categorisation of a contract (i.e. what sort of legal relationship it gives rise to), and what rights and obligations it creates, the court must look primarily at the substance of the agreement, and not at the labels which the parties have chosen to put on it. Examples of the sort of cases in which this principle has been invoked are where the question is whether an agreement creates a licence or a tenancy, or whether someone is an employee or a self-employed contractor, or whether an agreement does or does not create a legal partnership.

In many such cases, the contract will contain a label or declaratory statement, which the drafter hopes will persuade the court that it creates one form of relationship or obligation, and not another. An example in a property context is Street v Mountford [1985] A.C. 809. The issue there was whether a written agreement for residential occupation created a tenancy or a licence. The distinction was of crucial relevance in relation to whether the occupier was entitled to security of tenure under the Rent Act 1977. The agreement contained a statement to the effect that “I understand and accept that a licence in the above form does not and is not intended to give me a tenancy protected under the Rent Acts”. The House of Lords held that as a matter of substance the agreement satisfied all the requirements of a tenancy, and therefore that in law it was a tenancy, notwithstanding the label.

In a graphic, if slightly puzzling, illustration of the principle, Lord Templeman said in the course of his speech that: “The manufacture of a five-pronged implement for manual digging results in a fork even if the manufacturer, unfamiliar with the English language, insists that he intended to make and has made a spade”. It may perhaps be asked exactly what sort of digging he can have had in mind here, for five-pronged garden forks are rare beasts indeed – although the same cannot perhaps be said of muck forks. Or hay forks. But whether or not technically accurate, Lord Templeman’s comment underlines the essential principle: If it looks like a duck, swims like a duck, and quacks like a duck, then it probably is a duck, even if someone has mis-labelled it a goose.

And so, from the general to the particular. In Jervis v Harris (1996), the court was concerned with a common-form clause conferring on the landlord the right to inspect the premises, serve on the tenant a notice to remedy any disrepair found to exist, carry out the repair works if the tenant failed to do so, and recover the cost of the works from the tenant on demand. The question was whether the tenant’s obligation to pay was in law an obligation to pay damages (in which case, the Leasehold Property (Repairs) Act 1938 – and also s.18 of the Landlord and Tenant Act 1927 – applied) or a debt.

That question was one on which judges in earlier cases had roundly disagreed. In Swallow Securities v Brand (1983), McNeill J held that the obligation on its proper interpretation was one to pay damages, and therefore that the 1938 Act applied. A year later,

Vinelott J declined to follow Swallow, and held that the obligation was one in debt (Hamilton v Martell Securities (1984). His decision was followed by Nourse J in Colchester Estates (Cardiff) v Carlton Industries (1986), and by HHJ Paul Baker QC in Elite Investments v T. I. Bainbridge (Silencers) (1986), both on the ground that where a point has been argued before two judges at first instance, and the second judge after full argument has declined to follow the first, another judge asked to decide the point ought to follow the second judge unless convinced they were wrong.

And thus, the position remained until Jervis v Harris. As is well-known, the Court of Appeal held in that case that the obligation in question was in truth an obligation in debt, not damages, and therefore that the 1938 Act did not apply. The court reached its decision as a matter of the substance of the provision. What triggered the obligation to pay was not the state of disrepair but the carrying out of the works by the landlord at its own cost. So, the landlord’s claim was not a claim for compensation for loss suffered by reason of the tenant’s failure to repair, but a claim for reimbursement of expenditure which the landlord incurred in order to avoid that loss. It could not sensibly be described as a claim for damages for breach of the repairing covenant, because that breach had already been remedied. In other words, the claim was like a claim for a service charge: it was a claim under a promise to pay for the cost of work carried out.

Unlike the form of clause often encountered, the clause in Jervis v Harris itself did not contain any label, such as for example a provision that the cost was payable “as liquidated damages” (although it is fair to say that the obligation to pay “on demand” was perhaps more consistent with the obligation being a debt). However, it seems pretty clear that it would not have mattered if it had. The question concerned the legal categorisation of the bargain to pay: was it debt or damages? That question could not be conclusively determined by reference to a label – any more than the mis-labelling of a five-pronged manure fork can turn it into a spade.

That does not mean that a label is to be disregarded entirely. In some cases, it may be relevant, or even a helpful pointer, in deciding the meaning and effect of the agreement. But what is clear is that it cannot be decisive, and a proper consideration of the substance of the agreement may show that it is false or even misleading.

As always, if you wish to comment or share your experiences, please feel free to contact me at [email protected]