As I said in my Intervention on the Minister, I wanted to welcome the Home Secretary's conversion on Clause 1 in the light of last week's debate. Indeed, when I saw him on Monday I was overwhelmingly grateful because during a wide-ranging discussion he accepted the argument that the drafting of clause 1 was too loose. It took us some time to tease out the implications last week, but the upshot of our debate was that the offence could be committed as a result of negligence. Mindful of the Amendment that I tabled last week, the Home Secretary accepted that the offence should only be committed intentionally or recklessly. I was pleased with that result, and I thanked the Home Secretary. However, when I looked at the draft of amendment No. 34, I discovered that it reads very well only until the end of subsection (1). Subsection (1A) says:

"For the purposes of this section the cases in which a person is to be taken as reckless as to whether a statement is likely to be understood as mentioned in subsection (A1) include any case in which he could not reasonably have failed to be aware of that likelihood."

I always understood that in the normal test of recklessness, the word was to be given its ordinary English meaning. The problem has arisen because for a period of 20 years, which I suspect was an aberration, as a result of the Court of Appeal judgment in Caldwell, it became possible for a person to act recklessly even though he never foresaw the risk of something occurring. In a sense it was an objective test. The jury could say, "We accept entirely that in doing what you did, you never foresaw the risk of something occurring, but we can say that you are guilty nevertheless, because when we look at the facts, we think we would have foreseen that the risk would occur."

That decision stood until last year, when the House of Lords examined it again in a case called R v. G. As the Minister knows, before the House of Lords looked at it, the test in Caldwell had come in for massive criticism over the years from both academic and judicial authorities. In R v. G the House of Lords looked at the entirety of it, and the upshot was that the authority of Caldwell was wholly overruled.

It is worth considering the reasons why the Law Lords decided to do that. They stated, first, that while the most obvious culpable state of mind was an intention to cause an injurious result, knowing disregard of an appreciated and unacceptable risk of causing an injurious result or a deliberate closing of the mind to such a risk would be readily accepted as culpable also. But it was not clearly blameworthy to do something involving a risk of injury to another if, for reasons other than self-induced intoxication, one genuinely did not perceive the risk.

Secondly, the Caldwell formulation was capable of leading to injustice. Thirdly, the criticism of Caldwell by academics, judges and practitioners was not to be ignored, and fourthly, it was clear that the Majority's interpretation of "reckless" in Caldwell had been a misinterpretation of Parliament's intention in the Bill at that time, which was to leave the essential law unchanged. Therefore the test of recklessness should return to what it had been historically for many years before the Caldwell decision took place.

I have a slight criticism, which is that it took me a phone call to the draftsmen in the Home Secretary's office to get finally an admission that it was true that they had put in not the current test of recklessness, but the old test of recklessness. I know the Government do not like the judiciary, and perhaps the House of Lords particularly, but that is a crazy course of action to adopt.

In her comments, the Minister said, "We've done it in other cases, like rape." Yes, I acknowledge that. When we considered changing to an objective test in rape cases, there was considerable debate in Committee. The reason we chose to do it was the nature of rape as an offence, its seriousness and all the surrounding circumstances. To equate rape with words used by an individual seems a very poor comparison. The equation given in the discussion in R v. G refers to criminal damage, which is a good example of the narrow boundary between negligence and recklessness.

Where somebody is to be criminalised for something they said, on the basis that it might indirectly incite terrorism, we would be in danger of perpetrating injustice unless we said that the test that a jury has to apply is the current test, not the old test. The Minister will have to provide a compelling justification for me to be willing to accept the amendment. I questioned the draftsmen about the amended version. I do not think there is any difference between amendment No. 34 and the original draft. If there is, it is minute.

The real criticism of the earlier draft is that it was incredibly sloppy, because the Minister—I hope that she forgives me for saying this—was attempting to conceal what the Government were trying to do, which is why we ended up with the opaque language that I complained about last week.

Amendment No. 34 is not opaque. It makes it clear what the Minister intends, which is an objective recklessness test to allow a jury to conclude that although somebody had not been reckless in their own mind, they should still be criminalised. I do not want to labour the point, but that is a bad test to apply in this context.