Some areas of law are perennially unsatisfactory because the law has never spelt out clearly the legal consequences of everyday situations.

One area with more than an air of this is that of 'strangers intermeddling with trusts'.

This is an area traditionally dominated by the very different concepts of 'knowing assistance' (lately dead) and 'knowing receipt and dealing' (still with us).The problem is this: what is the legal liability of someone, a bank, for example, which allows trust money to be misapplied? A new answer was boldly provided on 24 May by the judicial committee of the Privy Council in Royal Brunei Airlines Sdn.Bhd v Philip Tan Kok Ming [1995] 3 All ER 97.

Baldly stated the answer is: nothing, unless that person can be shown to have dishonestly assisted in or procured a b reach of trust or to have knowingly received misapplied trust property.The defendant, Mr Tan, was the principal director of and shareholder in Borneo Leisure Travel (BLT), a Brunei-incorporated travel agency.

Unfortunately, BLT went insolvent when owing the airline the proceeds of certain ticket sales.

According to a standard form agreement, BLT held that money on trust for the airline.

However, in spite of the trust, it was BLT's usual practice to pay some of the money into its own bank account under a standing order, and to use that money for its own business purposes.

The judicial committee of the Privy Council, in an advice delivered by Lord Nicholls, held Mr Tan liable as a constructive trustee for the airline of the money wrongly paid into BLT's private account.The case concerned a travel agency and an airline.

Imagine, however, the facts of another mundane situation.

A cheque for £5000 has been dishonestly drawn on the bank account of a large charity by its treasurer.

The bank allows it to be credited to the treasurer's personal account, which is already in credit.

Is the bank liable as a constructive trustee of the £5000? Because of Royal Brunei Airlines v Tan, the answer given on 23 May would have differed in a number of respects from the answer given on 25 May.On 23 May, the analysis of the charity's advisers would have gone something like this: these facts probably disclose something called knowing assistance.

What do we need to show? We need to show (1) a trust or other fiduciary relationship; (2) a dishonest and fraudulent design on the trustee's part; (3) the assistance of the bank in that design; and (4) the knowledge of the bank.

How do we show (2)? Well, (2) is the taking of a risk to the prejudice of another's rights, which risk is known to be one which there is no right to take (see Peter Gibson J in Baden v Soci--t-- G--n--rale [1993] 1 WLR 509n).

How do we show (4)? Well, (4) is certainly shown if the evidence is that the bank either (a) actually knew of the existence of (2); or (b) wilfully shut its eyes to the obvious; or (c) wilfully and recklessly failed to make such enquiries as an honest and reasonable man would make.

However, if the evidence cannot support any of these, some authorities suggest that it may be enough to show 'constructive knowledge'.

For instance, where the bank (d) had knowledge of circumstances which would indicate the facts to an honest and reasonable man; or (e) had knowledge of circumstances which would put an honest and reasonable man on inquiry.

However, the weight of authority on 23 May would have strongly favoured an approach restricting 'knowledge' to types of knowledge which indicated dishonesty ('want of probity') on the part of the accessory, thus probably only the 'naughty knowledge' of types (a), (b) and (c) would have sufficed.

So what changes occurred on 24 May? Certainly there still needs to be a trust or fiduciary relationship (see (1)).

It is clear too that there must be the fact of assistance, subject to one point (see below).

The big change is that 'knowing assistance in a dishonest and fraudulent design' is dead.

In its place is dishonest procurement of or assistance in a breach of trust.

What then has become of (2) and (4)? Lord Nicholls puts it very simply.

As to (2), it does not matter what the state of mind was of the trustee originally in breach -- a breach of trust is a breach of trust.

On 25 May, therefore, (2) has become a breach of trust by the original trustee, regardless of whether the breach was innocent or fraudulent.

What of (4)? Accordi ng to the judicial committee the five levels of knowledge are 'best forgotten'.

Lord Nicholls emphasised that dishonesty should be the touchstone of the accessory's liability.

Mere knowledge should not be and is not now enough.

Thus it is that knowing assistance becomes dishonest assistance.

So how do you know when someone is dishonest in this context? Well, someone is dishonest who does not act 'as an honest person would in the circumstances'.

'In most situations', says Lord Nicholls, 'there is little difficulty in identifying how an honest person would behave.' It is an objective test and a pragmatic one.

However, Mr Tan was dishonest under this test because 'he caused or permitted his company to apply the money in a way he knew was not authorised by the trust of which the company was trustee'.

So the judicial committee accepts that states of knowledge still have a bearing on the presence or absence of dishonesty, as surely they must.If the bank, in our example, had been reckless, the circumstances might well have suggested that the bank had also acted dishonestly.

But what if the bank had been merely negligent? Although 'there may be cases where, in the light of the particular facts', a duty of care would be imposed on the bank, as a general principle, says Lord Nicholls, why should the charity have a remedy against the bank when it also has one against the trustees? An alternative which Lord Nicholls did not explore was the possibility of basing the accessory's liability in a tort other than the tort of negligence; there are sound arguments in favour of creating tortious rather than trustee liability for this wrong.

(Gary Watt considers this in greater depth in the Nottingham Law Journal, vol.1, 1995.)Nevertheless, it seems that something fairly radical happened on 24 May.

Something which, having its basis in principle rather than the confusion of the decided cases, has attempted to clarify the law.

Has it succeeded, though? A number of concluding points should be made.

First, it may be that Lord Nicholls' reasoning is limited to the particular context of 'commercial transactions'.

The opening words of his judgment suggest as much: 'The proper role of equity in commercial transactions is a topical question.' His Lordship also approves, inter alia, Knox J's description of an accessory as someone who is 'guilty of commercially unacceptable conduct in the particular context involved' (Cowan de Groot v Eagle Trust [1992] 4 All ER 700.

See also Eagle Trust plc v SBC Securities [1992] 4 All ER 488).

Secondly, demonstrating that somebody has acted dishonestly by Lord Nicholls' objective test may be problematic.

If, as his Lordship suggests, the test should take account of 'the personal attributes of the third party, such as his experience and intelligence, and the reason why he acted as he did'.

It is curious, also, that Lord Nicholls' test of dishonesty is different from that which is used to establish theft.Thirdly, Lord Nicholls suggests a uniform test for liability whether the accessory had procured the breach or merely assisted in it.

A uniform test may be almost inevitable in a corporate context, such as that in Royal Brunei Airlines v Tan, because when a director assists in a breach by his or her company he or she will very often have procured that breach also.

However, cases of procurement and assistance should perhaps be subject to different tests in non-corporate settings.

Fourthly, the judicial committee resisted the temptation to add to the jumble of judicial obiter dicta relating to liability based upon 'knowing receipt'.

Liability based on assistance/procurement is theoretically quite distinct from liability based on knowing receipt of misapplied trust property.

In practice, however, the vast theoretical differences are hard to justify.

Assume for a moment that the charity treasurer's private account with the bank was not 'in credit', but was overdrawn, the bank is now likely to be treated as having 'received' property in breach of the trust, thus opening the way for liability for knowing receipt, as opposed to knowing assistance.

Finally, but by no means the least important consideration, the case reminds us that the range of possible defendants in corporate insolvency situations might include the director, shareholder, employee or other agent who had been an accessory to a breach of trust by the company.

The sobering thought is that solicitors will continue to be desirable defendants and will presumably feel the full rigour of Lord Nicholls' objective test of dishonesty.There is, therefore, now a reformulated basis of liability for someone who dishonestly assists in or procures a breach of trust.

And much needed it was.

Before 24 May the accessory to a mere breach of trust could not be fixed with liability as a constructive trustee.

He or she would only be liable if the principal had been dishonest.

At least now the courts will be looking at the defendant's honesty or lack of it.

However, as we have observed, some problems remain with the test of dishonesty which has been chosen.

Perhaps another problem is the continued expansion of constructive trusteeship into areas which might be better regulated by the common law of torts.

The judicial committee in the recent case discussed a single form of liability in a specific context.

The general state of the law in this area remains confused.

As things stand, who knows when you might next be in a position of trust?