Nature and Evidentiary Value of Expert Evidence under the Bharatiya Sakshya Adhiniyam, 2023: Advisory or Conclusive?

The question is asked in every examination and answered the same way every time: expert evidence is advisory. The expert deposes and does not decide. But the proposition is easy to state and hard to apply, because in practice a court asked to determine whether two signatures were written by one hand, or whether a bullet came from a particular barrel, has no independent means of forming a view at all. Understanding what 'advisory' really requires of a court is the whole of this subject.

1. The Expert as a Witness

Section 39(1), BSA

When the Court has to form an opinion upon a point of foreign law or of science or art, or as to identity of handwriting or finger impressions, the opinions upon that point of persons specially skilled in such foreign law, science or art, or in questions as to identity of handwriting or finger impressions, or in any other field, are relevant facts. Such persons are called experts.

An expert differs from an ordinary witness in a way that is fundamental and is the source of every rule that follows. An ordinary witness deposes to what he perceived, and the court draws the inference. An expert deposes to an inference he has himself drawn, and asks the court to adopt it. He is not, in the language of the decisions, a witness of fact.

Two consequences flow immediately. The first is that an expert's evidence can only be tested by examining his reasoning, because there is no perception to probe. The second is that his conclusion, being an inference, is the very thing the court is supposed to be producing, and a court that simply adopts it has handed over its function.

The law resolves this not by excluding expert evidence but by insisting that the expert supply the material from which the court can reach its own conclusion. That is what 'advisory' means in practice, and it is why the requirement of reasons is not a formality.

2. The Governing Statement

📖 Ramesh Chandra Agrawal v. Regency Hospital Ltd., (2009) 9 SCC 709

Held: Where the court has to form an opinion on a subject requiring special skill, the opinion of a person specially skilled is relevant. But the credibility of such a witness depends on the reasons stated in support of his conclusions and the data and material furnished which form the basis of his conclusions. The court is not bound by the evidence of experts, which is to a large extent advisory in nature. An expert deposes and does not decide, and his duty is to furnish the judge with the necessary scientific criteria for testing the accuracy of his conclusions so as to enable the judge to form his own independent judgment.

Ratio: The expert supplies the criteria; the court forms the judgment. The credibility of the opinion rests on the reasons and data, not on the standing of the witness.

📖 State of Himachal Pradesh v. Jai Lal, (1999) 7 SCC 280

Held: An expert is not a witness of fact and his evidence is really of an advisory character. The duty of an expert witness is to furnish the judge with the necessary scientific criteria for testing the accuracy of the conclusions, so as to enable the judge to form his own independent judgment by the application of these criteria to the facts proved in evidence. An expert who merely states his conclusion without giving reasons or the data on which it rests is of no assistance, and his evidence cannot be acted upon. The court must satisfy itself of the expert's competence and of the soundness of his reasoning.

Ratio: A bare conclusion is not expert evidence at all. Reasons and data are what convert an assertion into assistance.

The two decisions together set out the whole of the doctrine, and every subsequent case applies them. Three requirements emerge — competence, reasons, and independent assessment by the court — and it is convenient to take them in turn.

3. Competence Must Be Established First

Section 39 admits the opinion of a person 'specially skilled'. Whether the witness is specially skilled is a preliminary question, and the court must be satisfied of it before the opinion is received at all.

The foundation is laid in examination-in-chief, by taking the witness through his qualifications, his training, the length and nature of his practical experience, the number of comparable examinations he has conducted, and any published work or recognition. What is required is not a degree — the section says nothing about qualifications — but demonstrated familiarity with the subject acquired by study, training or practice.

This step is skipped with surprising regularity. A report is exhibited, the author is examined about his findings, and nobody asks how he came to be competent to make them. The consequence is that the opinion is received without a foundation, and where the point is taken in argument the evidence is worth nothing, however sound the science behind it may have been.

⚠ Competence is field-specific

A witness is an expert in the field in which he is skilled and in no other. A medical officer competent to speak to the cause of death is not for that reason competent to give an opinion on the trajectory of a bullet, and a forensic scientist qualified in chemical analysis is not qualified in cyber forensics. The residuary words 'or in any other field' in Section 39(1) widen the categories of expertise the section recognises; they do not widen any individual expert's competence beyond what he can demonstrate.

4. Reasons and Data

An expert's conclusion is worth exactly as much as the reasoning that supports it. The Adhiniyam makes this explicit in two provisions that are easily overlooked.

Section 45, BSA — Grounds of opinion, when relevant

Whenever the opinion of any living person is relevant, the grounds on which such opinion is based are also relevant.

Section 40, BSA — Facts bearing upon opinions of experts

Facts, not otherwise relevant, are relevant if they support or are inconsistent with the opinions of experts, when such opinions are relevant.

Section 45 makes the reasoning relevant so that it may be placed before the court. Section 40 makes relevant facts that would otherwise have no place in the case, but which bear on whether the opinion can be accepted — published literature contradicting it, the results of other examinations of the same material, the conditions in which the examination was conducted, the adequacy of the sample, the accepted standards of the discipline, and the expert's own earlier opinions in comparable cases.

Together they constitute the machinery by which an opinion is tested. Where an expert states only a conclusion, that machinery has nothing to work on. This is why Jai Lal holds that such evidence cannot be acted upon: it is not that the conclusion is wrong, but that there is no means of finding out whether it is right.

What a properly reasoned opinion contains is, in substance, four things. What was examined, described sufficiently for it to be identified. How it was examined, including the method, the equipment and the conditions. What was found, set out as observation rather than conclusion. And why those findings support the conclusion, by reference to the criteria the discipline recognises.

5. The Court's Own Assessment

The proposition that a court is not bound by an expert opinion is stated in every decision, and it means something more demanding than a power to disagree. It means the court must actually evaluate.

A judgment that recites an expert's conclusion and adopts it, without examining the reasoning, has done what the doctrine forbids. Equally, a judgment that rejects an expert opinion without engaging with the reasons has failed in the opposite direction. What is required in both cases is that the court apply the criteria the expert has supplied to the facts proved in evidence, and explain the result.

The practical difficulty is obvious and is worth acknowledging rather than pretending away. A judge is not a ballistician or a molecular biologist, and where the science is genuinely technical his independent judgment is necessarily limited. What he can assess is whether the expert was competent, whether the method used is the accepted one, whether the data support the conclusion drawn, whether alternative explanations were considered and excluded, and whether the opinion survived cross-examination. That is the assessment the law expects, and it is a real one.

📖 Malay Kumar Ganguly v. Sukumar Mukherjee, (2009) 9 SCC 221

Held: Expert evidence is a rule of evidence and not a rule of law, and the court is not bound to accept it. The court must examine whether the opinion is founded on adequate material and sound reasoning. Textbooks and medical literature may be looked into for the purpose of appreciating the opinion, and the court may reach a conclusion different from that of the expert where the reasoning does not withstand scrutiny. At the same time, a court should be slow to substitute its own opinion for that of a qualified professional on a matter squarely within his discipline, where the opinion is supported by accepted practice.

Ratio: The court evaluates rather than defers, but the evaluation is of the reasoning, not a substitution of the judge's own view of the science.

6. Can a Case Rest on Expert Evidence Alone?

There is no rule of law forbidding it. The rule that emerges from the decisions is field-specific, and it turns on how far the discipline has developed and how much room the particular examination leaves for error.

At one end are disciplines the courts treat as reliable enough to act upon. DNA profiling conducted competently and with an unbroken chain of custody has been treated as capable of displacing even a conclusive statutory presumption. Fingerprint comparison, where an adequate number of ridge characteristics match and no unexplained difference exists, is long-established and is acted upon.

At the other end is handwriting comparison, which the decisions have treated with sustained caution. And in the middle sits medical evidence, which is ordinarily corroborative and rarely founds a case by itself, not because the science is doubtful but because a medical opinion normally establishes possibility rather than identity.

📖 Murari Lal v. State of Madhya Pradesh, (1980) 1 SCC 704

Held: There is no rule of law, nor any rule of prudence which has crystallised into a rule of law, that the opinion of a handwriting expert must never be acted upon unless substantially corroborated. But the imperfections of the science and the risk of error are such that the approach should be one of caution, and reasons for the opinion must be carefully probed and examined. Where the reasons are convincing and there is no reliable evidence throwing doubt on them, there is no reason to distrust the opinion. The hazard in accepting such opinion is not because it is unreliable but because human judgment is fallible.

Ratio: No absolute rule against acting on expert opinion alone, but the weaker the science the greater the caution.

📖 Magan Bihari Lal v. State of Punjab, (1977) 2 SCC 210

Held: It is unsafe to base a conviction solely on the opinion of a handwriting expert. Such opinion evidence is of a frail character and is hedged about with a great many uncertainties. Before acting upon it, the court must seek independent and reliable corroboration.

Ratio: The classical statement of caution, read in practice alongside Murari Lal, which declines to convert it into a rule of law.

The distinction that reconciles the two decisions is between a rule and a practice. There is no rule that expert evidence requires corroboration. There is a settled practice of looking for it where the discipline is one in which honest experts frequently differ, and handwriting is the paradigm of such a discipline.

7. Conflicting Experts

Where two qualified experts reach opposite conclusions on the same material, the court cannot resolve the conflict by counting heads or by preferring the more eminent witness. It must return to the reasoning.

The matters the court examines are, in substance, these. Whether each expert examined the same material, since a difference in what was examined explains a difference in conclusion without either being wrong. Whether the methods used were the same, and if not, which is the accepted method in the discipline. Whether either opinion rests on an assumption that the evidence does not bear out. How each expert fared in cross-examination, and in particular whether either was driven to concede that his conclusion admitted of alternatives. And whether either opinion is consistent with the other evidence in the case, which is frequently the decisive consideration.

Where the conflict cannot be resolved on any of these grounds, the position is that the fact sought to be proved by the expert evidence is not proved. In a criminal case that means the benefit goes to the accused, since a genuine and unresolved conflict between competent experts is itself a reasonable doubt.

⚠ The partisan expert

An expert retained by a party is under the same duty to the court as any other witness, and the courts have not hesitated to say so. An opinion that reads as advocacy, that omits findings adverse to the party retaining the expert, or that declines to concede the obvious in cross-examination, is discounted for that reason and not merely on its merits. Where an expert is shown to have deliberately given a false or slanted report, the consequences extend beyond the case, and courts have directed action against such witnesses.

8. The Report Is Not the Evidence

A recurring error is to treat an expert's report as though its production put its contents in evidence. It does not. The report is a document; what makes its contents evidence is the expert's testimony, given on oath and subject to cross-examination.

Three qualifications apply. Certain classes of report — those of specified government scientific experts — may by statute be used in evidence without the author being called, subject to the court's power to summon him. Where the maker of a report has died, the report may become relevant under Section 26(b) as a statement made in the discharge of professional duty. And a party may consent to a report going in without examination, though a party who does so cannot afterwards complain that the opinion was untested.

Outside these situations the expert must be produced. A report exhibited without its author is an untested opinion by a person whose competence has not been established, and it establishes nothing.

9. Expert Opinion and the Burden of Proof

An expert opinion never shifts the burden of proof, and it never dispenses with the proof of the facts on which it operates.

A forensic report identifying a substance as a narcotic proves what the substance examined was. It does not prove that the substance examined is the substance seized from the accused, and that must be established by evidence of the seizure, the sealing, the custody and the transmission. A report matching a bullet to a weapon proves the match; it does not prove that the weapon was in the accused's hands. A DNA report establishing a profile proves the profile; it does not prove that the sample came from the person alleged.

This is where a large proportion of forensic cases are actually lost, and the failure has nothing to do with the science. A favourable report on an article whose provenance has not been established is worth nothing at all, and the chain of custody is therefore the first thing a defence will attack and the last thing a prosecution should overlook.

10. The Position Stated Shortly

An expert is not a witness of fact. He deposes to an inference, and his function is to equip the court to draw its own.
Expert evidence is advisory, not conclusive. The court is not bound by it, and a judgment that adopts a conclusion without examining the reasoning has abdicated its function.
Competence must be established before the opinion is received, and it is field-specific.
A bare conclusion is not expert evidence. Reasons and the data behind them are what make an opinion usable, and Sections 40 and 45 exist to bring them on record.
There is no rule requiring corroboration, but there is a settled practice of looking for it where the discipline is one in which competent experts frequently differ.
Conflicts between experts are resolved on the reasoning, not by eminence or number, and an unresolved conflict means the fact is not proved.
The report is not the evidence. The expert must ordinarily be examined, subject to the statutory exceptions.
Expert opinion never shifts the burden, and the facts on which it operates — seizure, sealing, custody, identity of the sample — must be independently proved.

11. Related Topics and Provisions

Topic or provision	Connection
Expert and Opinion Evidence — Sections 39 to 45	The statutory scheme in full
Medical Expert Evidence and Ocular Evidence	The most frequently encountered application of these principles
Post-Mortem Report and Medical Evidence	The status of a report, and when the author must be examined
Substantive Evidence and Corroborative Evidence	The corroborative character of expert opinion in most fields
Section 40, BSA	Facts supporting or inconsistent with an expert opinion
Section 45, BSA	Grounds of opinion, when relevant
Section 26(b), BSA	Statements made in the discharge of professional duty, where the maker has died
Sections 57, 61 and 63 and the Schedule, BSA	Proof of the material a forensic expert examined, where it is electronic