When Oxford University's pharmacology department developed cracks in the plaster it sued the architect. So the architect sued the contractor – and lost. And thereby hangs a cautionary tale.
I bet every architect will sympathise with the Architects Design Partnership when it got sued by its client, Oxford University. The university said that the work done by Wimpey Construction (Now Tarmac Construction Contracts) was defective, but it didn't intend to sue the contractor. Instead, it wanted compensation from the architect.

Let me tell you what happened. Oxford University (the employer) decided to spend £9m to house its department of pharmacology. That was in 1989. The allegation by the university was that certain blockwork partition walls were not protected from the weather, and they became too wet. Once dried out, there was widespread cracking to the plaster finish.

Anyway, by mid-1991 the job was complete. There followed the defects liability period, which included repairs, to the cracking plaster and a long list of other snags. It went on through 1994, by which time all was in order. So the architect issued its "certificate of making good defects". Then, quite correctly, the final certificate flowed. So did the final wodge of money to Wimpey.

But by mid-1997, the university had trouble. More cracking plaster. It turned to its architect and complained that the architect had been in breach of its contract to inspect the Wimpey blockwork properly, had failed to instruct Wimpey to rectify the blockwork, had negligently certified practical completion then failed to see the remedial work was done properly, had failed to investigate the causes of cracking, had negligently issued the certificate of making good defects, had failed to recommend enforcement of the contract, then negligently issued the final certificate.


None of this was proved at that stage. But the architect figured that if the blockwork was as duff as the client said, Wimpey ought to jolly well pay up. So, it issued proceedings calling for the builder to pay the architect all the costs the court might find it liable for.

Do you see why the architect took hold of Wimpey by the scruff of the neck, shield and all, and dragged it into the action?

Now all this sounds a tad clumsy. If – and I repeat, if – Wimpey did make a mistake about the blockwork and plastering, why didn't the employer pursue the builder direct? Surely the true and direct cause of the loss was the alleged failure to build properly. That might be true, but had it gone to trial, Wimpey would have had a mighty shield. That thing called the "final certificate" under JCT80. It is conclusive evidence that the work is completed to the satisfaction of the architect, and also that the contract has been observed. So, do you see why the employer could not tackle Wimpey? And do you see why it says its architect was wrong to accept the plastered blockwork? And do you also see why the architect took hold of Wimpey by the scruff of the neck, shield, sword and all, and dragged it into the action? The technical point in all this is that the certificate operates as an evidential bar that precludes the employer from establishing that the work is bad and proving liability. That is all very well between employer and contractor, says the architect, but there is no evidential bar on us.

So it went to court. "A conclusion that a contractor might be liable to contribute would be wholly unacceptable," said the judge. He ruled that the final certificate under this JCT80 contract, if not questioned within 28 days, operates as a decision by the architect that the work is in accordance with the contract. The issue is tantamount to a decision discharging the contractor's liability. It is linked to the settlement of the final accounts, and thus of claims that each party might have had against each other.

So, if it is proved that the architect was negligent in its supervision of the contractor, knowing that the employer would have to fork out to put right the defects, the architect cannot then claim from Wimpey as though the negligence had not occurred. "That is rank injustice," said the judge.

The implication here, it seems to me, is that if the contract document lays down a piece of decision-making machinery for certificates, the certifier has to certify properly. If it does it negligently, then the cause of the loss to the employer is not the bad work but the issue of the certificate. So Wimpey may well have been let off the hook – rumour has it there may be an appeal.

One final point: JCT98 has wording that is different again. Take care.