Monday, August 16, 2010

Fire without smoke...

When David was taken to hospital, he was given an antibiotic injection which reacted badly with the insulin he took to control his diabetes. He suffered a hyperglycemic episode during which he pushed his way into a laboratory, knocking over a Bunsen burner. This caused a fire which spread quickly and burnt down one wing of the hospital. Everybody was safely evacuated and nobody was hurt.

Consider the criminal liability of David.

Thought points suggested:
If David was charged with arson, upon what would a successful plea of automatism depend? If his plea for automatism was accepted would it amount to sane or insane automatism? Why? How are the distinguished? Is there any question of self-induced automatism in this case? If so what are the rules of this vis-à-vis criminal liability?

If criminal liability is brought against David, then he is most likely to be charged with arson. Arson is a crime of criminal damage by fire: sections 1(1) and 1(3) of the Criminal Damage Act 1971. This offence is committed if by fire; D intentionally or recklessly, and without lawful excuse, destroys or damages property belonging to another. The prosecution could charge that David committed arson when he knocked over the Bunsen burner. They would seek to prove that his pushing into the laboratory was deliberate and argue that knocking over the Bunsen burner was at least reckless based upon his prior deliberate actions. A claim of aggravated arson would rest upon David’s intentionally setting the fire, in this case, as he knocked over a Bunsen burner, the prosecution would most likely seek to an offence of basic intent with the arson as a result of David’s recklessness.

The actus reus consists of damaging or destroying property by fire and clearly David did so.

The mens rea for arson comprises two elements: firstly that D recklessly damaged property, and secondly that he was reckless as to endangering life. In this context, recklessness bears a ‘subjective’ meaning; that is, the prosecution must prove that David was aware of a risk that life might be endangered and it was, in the circumstances known to him, unreasonable to take that risk (R v G [2003], overruling Metropolitan Police Commissioner v Caldwell [1981]). Note also Cooper [2004] in which the Court of Appeal held, applying R v G to a case of arson being reckless as to whether life would be endangered, that if D had been aware that there was a risk but had dismissed it as a negligible risk then D is not reckless.

David’s defence against the charge is most likely to be strongest if he asserts a lawful excuse – that of automatism (some may argue that automatism in fact absolves him of the necessary mens rea). However for a claim of automatism to fully succeed as a defence in absolving him of liability or possible hospital commitment, David should be able to claim that there was a total loss of control; that it was sane automatism, and that automatism was not based on some prior fault of his. He must seek to establish his innocence as Lord Denning outlined the defense in Bratty v Attorney General for Northern Ireland (1961).

David’s total loss of control can be established by medical facts as he entered a hyperglycemic episode. Such episodes have been successful used in cases to support a claim of automatism such as in Hennessy (1989).

However hyperglycemia is often related to insane automatism as it is seen as a disease of the mind (see the comments of Lord Lane in Hennessy (1989)). To qualify as sane automatism, David must therefore demonstrate to the court that his automatism as a result of the hyperglycemic episode was brought on by an external factor not related to the disease of the mind, something unique and quantifiable. David should be able to show that the antibiotic injection (which was indeed unusual and not in the normal course of events) was the external factor which caused his hyperglycemic episode and led to a loss of control. There is also a practical benefit to raising the defense as sane automatism versus insane automatism. The burden of proof of a claim of sane automatism once raised remains on the prosecution to prove the case beyond a reasonable doubt. While if a claim of insanity is raised by the defence, then the defence must prove it based upon a balance of probabilities. Thus for the defence it is easier to raise a reasonable doubt.

Turning to the last point, David should be able to demonstrate that the automatism was not caused by some prior fault of his. In crimes of basic intent, such as arson, which require only recklessness, prior fault of the defendant may be sufficient for a conviction if the defendant should have known the danger of their actions. The defence can reasonably claim that David did not have any prior fault in this instance. His actions in the laboratory were clearly related to the injection David received when he was in hospital. He was treated by staff of the hospital and given an antibiotic injection which reacted badly with the insulin he took to control his diabetes. David could argue convincingly that such treatment was not his fault (one would imagine that the decision was taken by the doctors based on David’s medical history, indeed one might suggest that the staff giving him the injection should have been aware of such a possibility to enter into a hyperglycemic reaction, however it is unlikely David would have been and as the injection was given by professionals he would likely be correct in following their advice.)

If David can raise reasonable doubt based on a claim of sane automatism supporting the notion of a loss of total control and with no prior fault of his own, then he should be cleared of the charge of arson likely to be laid against him.


  1. Hi Francis, if you'll allow me i'd like to make some comments on the essay. I haven't gone to much into depth due to time constraints, so i'll leave the criminal damage bit for now and concentrate on the lawful excuse of automatism, my immediate thoughts are as follows....

    You've covered the points well. Your mention of Bratty is a good one. Denning's commments of a mental disorder being a disease of the mind, that's prone to recur. Maybe the M'Naughten (1843) rules should have been mentioned?... As Bratty is a built upon the M'Naughten case?

    I'm also not sure if Hennessy would be the only case to use, as with him he didn't take his proper dose of insulin, rather than the external factor mentioned with David. Maybe other cases should be mentioned?

    When you mention the defence of sane automatism, a case should be used to support your point.... R v Quick (1973) and the apparent inconsistency between the Quick and Hennessy, clarified in Bingham (1991)

    Quick's defence resulted in one of sane automatism due to external factors. As david's behaviour was the result of an external factor, maybe this should be the case to mention and stress?

    There is also a distinction between hyperglaecemia and hypoglycaemia which perhaps should have been discussed, as Quick suffered from hypo, rather than hyper.

    Personally i think even though Quick's was hypo and David's was hyper, the principle of Quick could be applied to David, namely that an external factor produced his behaviour resulting in sane automatism, whereas Hennessy's resulted in insane automatism, because of an internal factor.

    I've found out my essay to send to you. It's handwritten, so i'll get a copy and send it as a pdf if you could give me couple of days.

    Paul Dale

  2. On that note, is the question definately stating hyper and not hypo? As the antibiotic injection reacted badly with his insulin, wouldn't it produce a hypo reaction, rather than hyper??

  3. Hi Paul,

    First of all a big thanks for the feedback. It means a lot to have someone look at the work and take the time to think it through and write the feedback. I am grateful for that.

    Secondly all the points you make are valid and good. It would strengthen the piece to mention M'Naughten, Quick, etc. The more I practice the more I realize that one should err on the side of giving a full picture of the case at hand.

    Finally, yes, oddly it states hyper (checked several times). Not what I would expect as hyper normally inclines to insane automatism rather than sane, I guess the point is to bring out the external factor issue.

    Thanks again for the help!