Deduction and Induction in the Courts

The two standards of proof are two settings of the same inductive dial. In a civil case the court asks which version is more probable. In a criminal case it asks whether guilt has been established beyond reasonable doubt, which is a demand for a much stronger inductive argument, not for a deductive one.

Where the deduction is

Once the facts are found, the application of the rule to them is deductive and admits of no discretion. If the rule is that a minor's agreement is void, and the finding is that this party was a minor, the agreement is void. A judge who accepted both and held otherwise would not be exercising a judgment; he would be contradicting himself.

This is why the conclusion of a judgment is usually a single sentence after twenty pages. The deductive step is the cheapest part of the whole process, and the entire cost lies in establishing the two premises.

Three devices that convert one into the other

The law of evidence contains machinery for turning an inductive inference into a deductive one, which is a striking thing for a logic student to notice, and MU's topic invites the noticing.

"May presume", section 2(1)(h): the Court may either regard the fact as proved unless and until it is disproved, or may call for proof of it. This leaves the inference inductive and merely permits it.

"Shall presume", section 2(1)(l): the Court shall regard the fact as proved unless and until it is disproved. The inference is now compulsory, but rebuttable: the inductive conclusion is imposed and may still be displaced by evidence.

"Conclusive proof", section 2(1)(b): on proof of one fact the Court shall regard the other as proved, and shall not allow evidence to be given for the purpose of disproving it.

That last one is the interesting case. By forbidding contrary evidence, the statute makes the step from the first fact to the second unchallengeable, and an unchallengeable step is a deductive one. Parliament has converted an inductive inference into a deductive rule by legislation. It does this where the value of certainty outweighs the risk of occasionally being wrong, and a student who understands the trade-off understands both the logic and the policy.

Where precedent fits

Following a precedent is not deduction. The earlier case is not a rule that contains this one; it is another dispute that resembles this one, and the reasoning runs: the earlier case resembles this one in the respects that mattered to the decision, so it should be decided the same way. That is argument by analogy, and analogy is inductive.

This is why distinguishing a case is a real argument and not an evasion. To distinguish is to say that a resemblance relied on is not relevant, or that a difference is. Those are exactly the tests for a good analogy, which are MU's own topic 1.9 and are worked in this module's last chapter.

It also explains why a line of cases is stronger than one case. More instances and greater variety strengthen an induction, and a proposition applied across many different fact situations is better supported than one applied once.

Where interpretation fits

Statutory interpretation is neither cleanly deductive nor cleanly inductive, and pretending otherwise is a common error.

Working out whether a bicycle is a "vehicle" within an Act is not a deduction from the definition, because if the definition settled it there would be no question. Nor is it an induction from instances. It is closer to what Module IV calls a précising definition: a decision, made for the purposes of this Act, about where to draw a boundary the ordinary word leaves fuzzy. That is why interpretation is taught in this course as its own subject and why the topic sits in Module IV of this one.

A worked example: one judgment, both kinds of reasoning

A shopkeeper is prosecuted for selling adulterated milk. The evidence is that a food inspector took a sample, that the sample was sealed in his presence, that the public analyst's report says the fat content was below the prescribed standard, and that the shopkeeper's signature appears on the seizure memo.

The inductive part. From the memo and the inspector's testimony the court concludes that this sample came from this shop, which no witness saw in its entirety and which is an inference from documents and testimony. From the analyst's report it concludes that the milk in the sealed container was below standard. From those two together it concludes that the milk sold was below standard. Each step goes beyond its evidence and each is judged by the prudent-person standard of section 2(1)(j).

The place where an inductive step is made compulsory. Where a statute provides that the analyst's report shall be evidence of the facts stated in it, an inference that the court would otherwise have to reason its way to is supplied by the legislature. That is the machinery of the previous section at work.

The deductive part. The rule is that selling an article of food below the prescribed standard is an offence. The finding is that the accused sold an article of food below the prescribed standard. The conclusion that he committed the offence follows necessarily, and nothing further is argued about it.

Where the appeal will go. An appeal against the finding that the sample came from this shop is an attack on an inductive conclusion, and it will be argued by pointing at gaps in the chain. An appeal against the holding that this substance is "food" within the Act is an attack on the major premise, and it will be argued out of the statute. The two appeals have nothing in common except the case they arise from, and confusing them is the surest way to lose both.

Distinctions that carry marks

What this does not mean

The syllogism is not a description of how judges reach decisions psychologically. It is a description of how a decision is justified. A judge may reach a view long before articulating the premises, as writers in the realist tradition have argued at length. The reply is that the justification is what is published, what binds later courts and what an appeal is against, so the syllogism remains the right account of the reasoning even if it is a poor account of the thinking.

Not every judgment is one syllogism. Most are chains of them, with the conclusion of one becoming a premise of the next, which is the chain argument of sequence 30.

Induction in a court is not guesswork. It is governed by rules of evidence, standards of proof and, in circumstantial cases, by a rule requiring every other hypothesis to be excluded. Those rules are the discipline that a bare inductive argument lacks.

Quick revision

Facts inductively, law deductively. That one line is the answer to the question as MU prints it.

Section 2(1)(j), BSA 2023 defines "proved" by a standard of probability and by what a prudent person would act on. An inductive standard in a statute.

Standards of proof: preponderance of probability in civil cases, beyond reasonable doubt in criminal ones. Two settings of one inductive dial, not a change of logical kind.

Presumptions: may presume, 2(1)(h); shall presume, 2(1)(l); conclusive proof, 2(1)(b), which forbids contrary evidence and thereby makes the step deductive.

Precedent is analogy, therefore inductive, therefore judged by the relevance of the resemblances. Distinguishing is an attack on the analogy.

The judicial syllogism describes justification, not psychology. That is the realists' criticism and its answer.

Test yourself

1. Explain the use of deductive and inductive reasoning in a court, with the stage of the case each belongs to.

Finding facts from evidence is inductive: no judge witnessed the events, so the court reasons from testimony, documents and circumstances to a conclusion that goes beyond them and is reached on a standard of probability. Applying the legal rule to the facts once found is deductive: given the rule and the finding, the result follows necessarily and no further evidence is relevant. Almost every argument in litigation is an argument about one premise or the other.

2. How does the Bharatiya Sakshya Adhiniyam 2023 define "proved", and why is the definition inductive?

Section 2(1)(j) provides that a fact is proved when, after considering the matters before it, the Court either believes it to exist or considers its existence so probable that a prudent man ought, in the circumstances of the particular case, to act on the supposition that it exists. It is inductive because it asks for probability and not for necessity, and because it fixes the required degree by reference to what a prudent person would act on rather than to any demonstration.

3. Distinguish "may presume", "shall presume" and "conclusive proof".

Under section 2(1)(h) the court may either regard the fact as proved unless disproved or call for proof of it, so the inference is permitted only. Under section 2(1)(l) the court must regard it as proved unless and until it is disproved, so the inference is compelled but remains rebuttable. Under section 2(1)(b) the court must regard it as proved and must not allow evidence to disprove it, so the inference cannot be challenged at all, which makes it deductive in effect.

4. Why is following a precedent an inductive process?

Because the earlier decision is not a rule that logically contains the present case; it is another dispute said to resemble this one in the respects that mattered. The reasoning runs from resemblance to like treatment, which is argument by analogy, and analogy is a form of induction. This is why distinguishing is a genuine argument: it denies the relevance of a resemblance or asserts the relevance of a difference, which are the standard tests of a good analogy.

5. Is statutory interpretation deductive?

No. If the words of the definition settled the question there would be no question to argue, so interpretation is not a deduction from the definition. Nor is it an induction from instances. It is a decision, made for the purposes of the particular Act, about where to draw a boundary that the ordinary word leaves indeterminate, which is what Module IV calls a précising definition.

6. What is the realist criticism of the judicial syllogism, and what is the answer to it?

That the syllogism describes how a decision is presented rather than how it is reached, judges often forming a view first and constructing the reasoning afterwards. The answer is that the syllogism is an account of justification, not of psychology, and justification is what matters institutionally: it is what is published, what binds later courts, what the parties can attack, and what an appellate court reviews. How the judge came to the view is not reviewable; the stated reasons are.