His Honour Adrian Head describes the function and duties of an expert witness

English law finds it convenient to define by categories. The phrase 'expert witness' identifies a distinct cog in our legal machinery. The trap is to think that the category not only defines but tells the whole truth about the subject: the category 'expert' includes the good and the bad; this is true of expert witnesses.

The expert witness should never be a party's advocate but a person who, having understood the parties' relevant allegations, can see whether they correctly define the issues to which his expertise is to be directed, and, pinpointing any discrepancies, can put his expertise impartially at the disposition of the judge to assist him to perform his task of rightly deciding an issue before him. A genuine readiness to accept another's view, if preferable, is essential in an expert; unqualified loyalty to one's own opinion is not acceptable. First, he must master the facts and their consequences alleged by the parties. In many personal injury cases this is simple because the expert is dealing only with the physical consequences of an acknowledged event: but, even then, important differences of alleged causation and consequence can arise.

Such differences are particularly relevant in building and engineering cases where more than one craftsman or supplier has contributed to the final article. Long ago, for example, in an unreported case, the undoubted failure of the ahead / astern control in a small converted vessel was blamed by the owner on defective manufacture and by the supplier on the owner's improper operation of it under power. Issue was joined without relating the fault to other equipment until an alert expert established that the error was not in manufacture nor in operation but in the shipwright's choice of a propeller inappropriate for that control.

Correct consideration of the parties' allegations, particularly in building cases, may well persuade a retained expert to advise his client that an additional party should be joined in the proceedings. (It is unclear how the single expert envisaged in the Access to Justice Report, advisor to the court not to a party, is to act in those circumstances).

Secondly, in preparing his report the expert must relate the perceived issues to three sources of information;-

1) Initially comes the examination of witness statements of parties and other witnesses,

2) Next, the expert must detail the relevant facts perceived by him on examination:

(a) of the locus in quo, when terrestrial features may have contributed to causation or have mitigated consequences; and (b) of the subject matter, whether damaged or defective bodies, minds, machines, premises or damage of a financial nature or to intellectual or other property of any kind.

In this task the expert must always avoid confusing assertion and truth, carefully distinguishing perception from deduction so that the court and other experts can see clearly the links in his evidential chain. Thus, 'I saw', 'I heard' are key words: phrases like 'I examined' merely stipulate compliance with the duty to enquire.

In cases of physical or mental impairment the patient's complaints and assertions are, as every good doctor knows, matters of fact but not necessarily true. If clinically important they should be reported: truthfulness as a clinical perception, as when reported complaints of pain on examination are accompanied by accounts of possible genuine involuntary and clinically relevant events such as shortened breath, pupil dilation and so forth, distinguishable from equally important voluntary manifestations ranging from slight exaggeration to lead-swinging demonstration.

Medical experts must always remember that their accounts of what strikes them as truthful are only ever evidence of their impressions; decisions as to truthfulness are for the court alone on the totality of the evidence including the test of the witness box. Thus the expert must always report on his examination as an observer and not as an advocate; although, in truth, when damage to personality or the brain's function is an issue, and communication is not simply a matter of report but may itself be a vital symptom of particular injury or disease, the boundaries between adjudication and observation are more difficult to define.

Furthermore, references to mechanical or electrical diagnosis of human or material injury, such as ultrasound or CT scan, should always be identified as interpretation, not observation, since, in using such methods, the expert does not himself observe but is usually dependent on readings obtained by the interpretive skill of the operator.

Reporting on damaged or defective machines or constructions, or failed industrial procedures or faulty accounting and finance is more nearly a purely objective process since the subject-matter has no voice: even here the expert must be alert, not without explanation, to express as a fact what is, in truth, observed fact plus opinion on it. The common phrase that a component 'was badly worn' is less a description than a judgment to be justified, in all but the simplest cases, by a description of an undamaged component, it's designated use and acceptable related measurable wear, and only thereafter by an account of actual wear defined by photo / measurement etc. and attribution of the cause.

Especially when human activity is concerned, experts need to be alert to the degree to which their stated observations involve judgments: so, in financial matters, while the spending or receipt of a sum of money may be an objective fact, comment on the propriety of it's appropriation or allocation in a set of accounts usually involves a subjective decision (fully to be acknowledged) as to the appropriate accounting practice and the correctness of it's application. Likewise, with translations the dictionary deceives: even to translate German vertrag and French contrat as English 'contract' can be a misleading simplicity. What English probation and welfare officers call a contract, although not one in English law, might to a Frenchman acquire validity from Art 1103 of the Code Civil .

Especially when faced with words of widely differing social or historical import, an interpreter can be the most dangerous of expert witnesses.

3) To prepare his report the expert must relate his knowledge of the issues and his perceptions and observations to what I would call Ôthe facts of assumption', namely:

(a) in the case of damage physical or mental to a person, or physical detriment to or defect in an article or building, etc, he must, unless it be agreed, assume what was the unaffected / undamaged condition of the relevant person or object before the litigated event, for only thus can the supervening damage be identified; and he must make and express that assumption as a matter of evidence, which may be tested like any other.

(b) in dealing with alleged errors in design, construction, accounting or translation the expert must prognosticate an assumed correct model and from that identify the errors in or conformity of the challenged example. The sources of his prognostication, however, are matters of evidence open to challenge.

In both cases (a) and (b) the expert has total freedom to choose the sources of his assumptions so long as they are disclosed, since their value and reliability are open to challenge.

Personal and professional qualification and experience are useful benchmarks provided that long service does not mask an ignorance of and unwillingness to face and absorb new ideas and analytical methods. Length of experience needs to be qualified by the professional and geographical areas in which it has been gained: thus the geographical value of a marine surveyor's experience of ice effect in the Baltic or of teredo worm in the Caribbean is more valuable than length of service to the owner of a relevantly affected wooden yacht. Comparative examples are legion: their significance is that in choosing an expert witness, or in assessing the quality of his evidence his provenance as an expert is itself most material evidence.

Sources of information from learned works are usually acceptable as to past facts. So long as the expert identifies the basis of his assumptions in sufficient detail to enable any appropriate challenge to be made he may elliptically treat them as statements of truth; for example, 'relying on (author) it appears that the chances against (say) failures of a big end in a well-lubricated diesel engine are x:1'. Similarly, in cases of injury and disease references may be made to scientific works; but when considering an aspect of a living organism - less easily photographed or inspected, the expert must be sure of accurate categorisation of the relevant condition to which the scientific literature truly relates.

The expert must correlate the three classes of information, namely of the facts alleged, the facts perceived, and of the facts assumed in order correctly to interpret the available information:

(i) to identify the relevant cause(s) of the events being litigated or their consequences; and / or (ii) to afford a reasoned prognosis whether for human or material injury, for damaged goods or financial well being; and / or (iii) by correct analysis to enable the court to determine rights which depend on language, whether it is the legalese of documents related to features of land or the imbalance of cultural, legal and philosophical concepts which dictionaries often fail to identify, even in the UK and USA.

Having so armed himself, the expert must still be ready to face the court where proper advocacy, probing judges and fresh evidence may cast doubt on careful observations and analyse every contested conclusion in pursuit of loopholes in logic and mistakes of fact. He must remind himself that he is the servant of the court, not of any litigant; that all his expertise affords him no didactic or judgmental voice but only the right to supply, if acceptable, the raw materials for someone else's decision. He must at all costs be ready to consider alternative views; and, when in conscience such view is to be preferred, at least to accept that possibility. His language should aim to be so simple as to be understood not only by country judges but by anyone involved in the litigation. He should not see himself as condescending to share esoteric information, but as a servant of justice performing a public duty.

The checklist for the expert witness should then be:

1. What truly are the allegations which concern me; do they define what I see as the true issue(s)?

2. Have I studied all the witness statements and disclosed documents to ascertain the facts?

3. Have I made all appropriate factual observations of bodies, places, machines, buildings, accounts, etc?

4. Have I specified for myself the assumptions that I shall rely on so clearly that, if challenged, I can detail them and identify all my sources?

5. Have I prepared my report in simple intelligible English, using as few technical words as I can, explaining those that necessarily remain?

6. What, if any, do I see as the points of challenge to my observations, assumptions and conclusions; and what, if any, are the answers to them?

7. Since the expert is not an advocate but must inform the mind of the court fully, not partially, have I improperly omitted anything relevant?

Adrian Head has recently retired from the circuit bench. Arrangements for his services as a mediator and arbitrator can be made through his old Chambers, 3/4 South Square, Gray's Inn, London WC1R 5HP. Tel. 0171 696 9900