Alternative dispute resolution (ADR) tends to divide commentators into true believers and total sceptics.

The former are found in high concentration among ADR professionals; the latter tend to have been less practically involved and often rest their case on philosophical objection.

So, steering a middle course, what can we say about the effectiveness of ADR?A platform from which to examine this subject is the United States.

Americans have adopted ADR with typical gusto.

Arbitrators and mediators have earned themselves academic support, legislative backing and sophisticated research programmes.

The twin early streams of community mediation and commercial arbitration have combined to provide a mighty Mississippi of various court-annexed ADR schemes.

The result is a welter of written material, much of it generated by the programme designers, organisers, trainers, monitors and evaluators of a host of different schemes.ADR has now decisively crossed the Atlantic.

The Legal Aid Board is proceeding with a mediation pilot scheme in the context of the new family legislation.

Lord Woolf plugs away at commending ADR.

So too does Lord Mackay in his legal aid green and white papers.Domestic experience, plus that of the United States of America and elsewhere, suggests that ADR, particularly mediation, works well between consenting commercial companies, divorcing couples and in disputes between neighbours.

The big question is whether ADR has any future in terms of incorporation into court proceedings.

The Lord Chancellor's Department has blown somewhat hot and cold on testing this but has now established a pilot in-court ADR scheme at the Inner London County Court.

However, take-up is reported to be somewhat low.While awaiting the results of these home-grown experiments, it might be worth looking at lessons emerging from the US.Since the late 1980s, the US Congress has had a growing interest in ADR.

A test group of a small number of courts has been established with in-house ADR schemes, half semi-compulsory where parties have to opt out and half voluntary where parties opt in.

Their effectiveness is being researched by two bodies: the RAND Corporation which is a private foundation and the Federal Judicial Ce nter which is similar to a souped up, and properly resourced, Judicial Studies Board.

Both bodies are close to final report.The early indications of this serious research can be compared with the rather more random empirical experience.

A somewhat sceptical and rights-orientated lawyer, I visited a number of ADR schemes in the US and Canada during the summer of 1995 for a project funded by the Nuffield Foundation which gave rise to the Legal Action Group's publication, Achieving Civil Justice.

It turns out that the official research will be somewhat ambiguous.

In such circumstances, personal evaluation takes on rather more value than it perhaps in theory should.

Findings suggested to me four major lessons.Firstly, it is evident and perhaps predictable that, in the US, ADR is still in its pre-institutionalised, charismatic phase.

Successful ADR schemes were headed by wonderful, enthusiastic, highly motivated and skilled people.

A team with personnel of this quality could succeed at most things, from ADR to rocket design.Secondly, ADR, and in particular mediation, has the potential to offer litigants power well beyond the normal court process.

Parties are involved and describe their version of events in a way impossible in the more formal surroundings of a court.Thirdly, ADR can be integrated into a court's process so settlements are made earlier.

ADR brings the settlement dates forward.

Sceptics argue this is simply the result of the lawyer having to work on the case to meet the demands of the ADR process.

This undoubtedly has some truth but lawyers in places with well established ADR schemes, such as Boston or San Francisco, agreed they had created a 'settlement culture' through a greater expectation that cases would settle earlier.This was good news for clients, who saved on legal fees; neutral news for courts, for whom it made relatively little difference; and bad news for lawyers, at least those who received lower fees but no greater volume of work.

However, contrary to cynical prediction, many lawyers liked ADR.

They responded positively to the idea of working out clients' problems collaboratively rather than confrontationally.Finally, and this is good news for solicitors, most North American ADR schemes welcome legal representation.

Indeed, some successful schemes, such as that in Cambridge, regarded litigants in person as prima facie unsuitable for ADR.

Schemes recognised that lawyers helped to correct any imbalance of power between the parties and the neutral mediator.So, what preliminary judgement can we make? Court-annexed ADR can 'add value' to the litigation process though it should be seen as alternative to courts, not lawyers.

ADR may, ultimately, save some parties some money in legal costs but is not an easy source of savings for the government in relation to the cost of legal aid or the court service.

Litigants in ADR processes need lawyers.

This perception knocks out ADR's most likely institutional backer in the United Kingdom.

Alas, that leaves it as a desirable initiative unlikely to be very widely implemented.