The common law, says Tony Weir, has never been inclined to split the difference and declare a draw.1 Its traditional preference is for all-or-nothing outcomes: full recovery or none at all, total defences, not partial ones. However, this polarized approach has increasingly been softened through statutory interventions, such as the power to reduce damages by a “just and equitable” amount in a tort action, to reflect the claimant’s contributory negligence.2 Mr Weir suggests that there has been some relaxation in the common law itself, citing the House of Lords’ award of a conventional sum for “loss of autonomy” to escape the dilemma of awarding either the full cost of bringing up a baby or nothing at all, in Rees v. Darlington Memorial Hospital NHS Trust .3 Yet, ever in touch with the Zeitgeist , Her Majesty’s Judges may since have concluded that Third Way “thinking” has seen its popularity decline in step with that of New Labour and its chief evangelist, Mr Blair. At any event, they have stopped the middle-way trend in its tracks in Gregg v. Scott ,4 with the House of Lords rejecting a claim for the reduced chance of recovering from cancer in a medical negligence case. The all-or-nothing approach to causation in personal injury cases seems to have been authoritatively re-established, at the cost of the “loss of a chance” analysis.

Mr Gregg went to Dr Scott concerned about a lump which had developed under his arm. Dr Scott negligently diagnosed it as benign, whereas it was in fact cancerous. The cancer continued to grow, spreading into Mr Gregg’s chest. His condition was diagnosed by a different doctor some nine months later. It was found that Mr Gregg’s chances of disease-free survival for 10 years were 42% at the time of the initial consultation with Dr Scott, and the delay had reduced this chance by some 20%. Prompt treatment would probably have achieved remission without the need for (traumatic) high dose chemotherapy, and would (at least initially) have prevented the cancer from spreading into the pectoral region.