UGC NET Logical Reasoning: Square of Opposition and Indian Logic

The official boundary: what belongs in Unit VI

The UGC syllabus divides this unit into several linked tasks. Formal logic asks whether a conclusion follows from a form, while informal fallacies ask whether an argument has a reasoning defect despite natural language. The square relates four categorical sentence forms about the same subject and predicate. Venn diagrams test whether premises force a conclusion. Indian logic studies means of knowledge and inference through its own terms. Do not assume that a question on one of these tasks gives the entire unit a larger weight; the syllabus note gives five questions of two marks from each of ten units.

NTA’s December 2026 bulletin says Paper I has 50 compulsory questions for 100 marks, Paper II has 100 questions for 200 marks, and both share a three-hour sitting without a break. Each correct response earns two marks and an incorrect response loses none. It provides no separate Unit VI clock or official question-type quota. Any time allocation in this workshop is a personal practice rule, not an NTA instruction. The current bulletin also says to choose the most appropriate option when there is no negative marking.

First translate the sentence: A, E, I or O

A and E are universal propositions; I and O are particular. Quality means affirmative or negative. Keep the same subject class S and predicate class P as you compare them. “All researchers are teachers” and “Some teachers are researchers” change the subject and predicate positions, so they are not simply two corners of one square. “Some” in this logic means at least one, not “some but not all.” That last distinction prevents a common wrong answer.

The two diagonal pairs are the easiest to apply: A and O cannot share a truth value, and E and I cannot share a truth value. If “All S are P” is false, “Some S are not P” is true; if it is true, O is false. If “No S are P” is false, “Some S are P” is true. These diagonal relations are contradictions. They do not depend on a separate guess about how common S is.

Three other relations, with the classical assumption stated

A and E are contraries: they cannot both be true if S has at least one member, but they can both be false when some S are P and some are not. I and O are subcontraries: they cannot both be false under that same nonempty assumption, but they can both be true. Subalternation goes from A true to I true, and E true to O true. It also goes from I false to A false, and O false to E false. Do not reverse the truth arrow: “Some S are P” does not prove “All S are P.”

Here is a quick self-test. Suppose “No members of this reading group use the library” is false. This is E false, so I is true: at least one member uses it. You cannot infer that every member uses it. If instead “Some members do not use the library” is false, O false makes A true: every member uses it. The two English sentences sound close, yet they force different conclusions. Try both in the checker, then explain the result without looking at the table.

Modern empty classes: a caution for advanced questions

The syllabus specifically says “classical square of opposition.” Traditional relations outside the diagonals assume there is at least one S. Modern Boolean logic does not treat “All S are P” as proving that an S exists: if there are no S, the universal can be true while “Some S are P” is false. If a question explicitly says the subject class may be empty, rely on the diagonal contradictions and read its stated interpretation. Many short notes omit this condition and turn the vertical arrows into unconditional facts. We include it so the diagram does not teach a false universal rule.

From two premises to a valid conclusion

A square question relates one categorical proposition to another. A syllogism adds a second premise. First identify the middle term that appears in both premises but disappears from the conclusion. Then look for a counterexample: can both premises be true while the proposed conclusion is false? If yes, the inference is invalid. A Venn diagram is a disciplined way to draw this test. Shade a region when the premises say it is empty; place an X when they say at least one member exists. Do not add an X for “all” unless existence is given.

The second and fourth examples expose an everyday trap: two premises can each sound supportive while their named people are different. Draw the minimum picture allowed by the premises, then try to place the “some” groups separately. If you can, the conclusion is not forced. This is more reliable than memorising a slogan such as “two particular premises are always wrong” without understanding which inference is being tested.

Deduction, induction and a five-minute counterexample test

A deductive claim promises necessity: if the premises hold, the conclusion must hold. The counterexample test is therefore decisive. An inductive claim offers support without necessity: several observed library users may suggest a pattern, but another group could behave differently. The same passage can contain both. “Every surveyed reader preferred a digital copy” is a fact about the sample; “all readers everywhere prefer digital copies” goes beyond it. In a Paper I question, mark the scope of the evidence before choosing “valid,” “strong” or “probable.”

Indian logic: six named means of knowledge

The UGC syllabus names six pramanas: pratyaksha, anumana, upamana, shabda, arthapatti and anupalabdhi. This is the syllabus list for the exam, not a claim that every Indian philosophical school accepts all six. For revision, ask how a claim was obtained. Was the object perceived, inferred through a dependable relation, learned by comparison, reported by a trustworthy speaker, postulated to explain a fact, or known through non-apprehension in appropriate conditions? Keep that route separate from whether the conclusion happens to be true.

These illustrations are deliberately simple. They are not official questions and should not be used as complete philosophical definitions. In particular, ordinary “I did not see it” is weak evidence of absence when lighting, access or attention is poor. Similarly, an inference needs a dependable vyapti, the invariable connection named by the syllabus, rather than a coincidental sign. A reported sentence depends on its source. This second check—conditions under which the route works—is what makes the list useful instead of six labels to memorise.

Anumana, vyapti and where an inference breaks

A standard classroom model of anumana runs from a subject through a reason or sign to a conclusion, supported by a general relation. “This hill has fire because it has smoke” is the familiar model, but only if the smoke–fire link is secure and the smoke really belongs to that hill. Our fresh example is a laboratory thermometer: if a calibrated instrument reliably reads above a defined threshold and it is measuring this vessel, an inference about the vessel’s temperature may be warranted. If the instrument belongs to another vessel, the reason is misplaced; if it is faulty, the general connection fails. These are two different failure points.

The syllabus names hetvabhasas, fallacies of inference. Instead of treating every ordinary mistake as one named subtype, first ask three diagnostic questions: Is the sign established in this subject? Is the sign reliably connected to the proposed conclusion? Is a contradictory observation or stronger reason present? These questions do not replace a full classical taxonomy; they stop a candidate from labelling a weak inference “proved” merely because the example resembles a memorised formula. Use a formal text for traditional subtype names if an exam item asks for them.

What six public Bureau booklets show, and what they cannot show

Our earlier question audit coded six publicly available UGC NET Bureau Paper I booklets from June 2015 to July 2018, totaling 340 questions. Within their Logical Reasoning blocks, we identified ten square-of-opposition items across the six booklets; Indian logic terms were not directly tested in that sample. The source file for the count is in the ProSyllabus research alongside the syllabus guide. The sample is old, limited, and predates NTA’s computer-based format. It supports practising the square as an observed historical question family, not skipping Indian logic today. Current NTA answer keys disclose question IDs and options without the public question text, so a defensible new Unit VI frequency chart cannot be computed from those keys.

A seven-session repair plan

Work on one reasoning error at a time. For each miss, write the premise, your conclusion, the counterexample you should have tried, and the corrected rule. Redo a fresh case rather than the same remembered answer. The board below saves check marks locally on this device; it does not collect a score or prove readiness. Spend at least one session on the named Indian-logic terms even if your older paper practice has few examples, because the current syllabus still includes them.

Questions candidates ask