Unlawful killings are invariably messy in the real world yet the criminal law likes clarity. This tension can reach fever pitch where two or more causes are involved. Appellate judges in such “novus actus” cases, naturally desiring justice given any set of facts, seem to me to go about their task backwards, deciding what the result should be then deriving the principles which must flow. Am I being too harsh? Or cynical?!

At the risk of sounding exactly like an academic essay question, of A to D, who is criminally responsible for death in the following scenarios?

(1)    A punches V who falls into the road, unconscious but with minor injuries. A runs off and, 20 seconds later B, a taxi driver, fails to see him and runs him over, killing him. What if the car came 10 minutes later? What if B was driving without any fault and could not avoid it?

(2)    Driver A pulls out of a junction without looking, as he is texting, hitting driver V, whose car is pushed onto the opposite carriageway so colliding with driver B, which second collision kills V.

(3)    A stabs V which causes him to collapse but the wound is not of itself fatal. B and C carry V on a stretcher for treatment but drop the stretcher on the way and then medical treatment by D is sub-standard, causing further injuries which in combination cause death.

(4)    A shoots at police, then grabs V as a human shield. Police officer B returns fire and shoots V dead. Police officer C is present also shooting but his shots miss. 

(5)    2 enemy gangsters A and B have a gun battle. A accidently shoots a passer by V, B also knowing that others such as V may be in the vicinity. 

(6)    A punches V unconscious at a beach and leaves the scene. B, a friend of A who had planned with him to rob V, then takes his wallet, moving V a little closer to the shoreline and then leaves. The waves sweep V away, who drowns. What if the wave were a freak tidal wave and he would not have ordinarily drowned in that position?

(7)    Finally, A supplies V with heroin in B’s presence. B knows V has been having suicidal thoughts. In front of both of them, V self-administers the heroin and dies from an overdose.

The simple starting point is the established proposition that if A’s act is unlawful, it need only be one among many causes of death, even not the main cause, so long as it is more than trivial, for A to be culpable. But when does a subsequent act break the chain of causation and make A not guilty of any homicide?

When you look at English law, there appear to be several answers to that question. Some authorities focus upon the degree of independence of the subsequent act. In other words, was the intervening cause so overwhelming as to consign the first cause to merely part of the background facts? In that approach, our stabber in (3) is still guilty as the acts of getting V treated were not independent of his stabbing. Our hostage taker in (4) is still guilty, as the police shooting back in execution of legal duty and in self-preservation was only done as they were being shot at and they never intended to kill the poor hostage.

Other appellate courts focus on the concept of foreseeability. So it was held that our puncher A in example (6) would not be guilty where he could not have foreseen that B might not only rob V but leave him nearer the water (or indeed try to drown him). This seems like a subjective test – asking what did A actually foresee? Yet this would allow A to escape conviction where he didn’t foresee what might happen even where it would be obvious to a reasonable person. Not surprisingly, other appellate decisions, both in the UK and Canada, have formulated this as an objective test – whether acts of the general type which later occurred could reasonably have been foreseen by A. If so, he remains guilty. This is the approach of the Court of Appeal in driving cases, such as example (2). Yet applying this test, could our stabber in example (3) reasonably foresee that the stretcher bearers would drop the patient and the surgeon screw up?

Confused!? Where is the overarching principle here? It doesn’t end there. When the House of Lords came to examine facts similar to (7) above – our heroin taker – it was clear that what V did was perfectly foreseeable to A and B and anyone in their position. Yet it was held the chain of causation was broken and A - the supplier - should not be responsible. The principle seemed to be that V made a voluntary and informed choice to self-administer knowing it was heroin, so A should not be criminally liable for causing his death. In a near confession of the muddled and varying approaches the courts adopt to these cases, the House of Lords observed that causation is not a single, unvarying concept to be mechanically applied without regard to the context in which the question arose. Pick the bones out of that!

One response might be to say that, where the subsequent cause of death is not the result of a deliberate act (eg a careful driver has a genuine accident running V over, negligent medical treatment of V), it is useful to focus on whether that act would have been reasonably foreseeable to A, in other words whether the subsequent act was so extraordinary or on the other hand often encountered. Where the subsequent cause was the result of a deliberate act of free will, such as another deliberate assault on V or V acting unlawfully of his own free will, it might be useful to focus on how independent the subsequent act was of the original act.

An overarching principle in this area seems to remain elusive. And none of this directly answers whether B, C or D should also be tried for their own conduct. I’m all for the law being flexibly applied but it is absolutely vital, whether prosecuting or defending these multiple cause cases, that early and transparent consideration is given to the correct formulation of principle and how the jury might be best assisted when considering their verdicts, in order that both victims and defendants can access justice.