Exceptive Syllogism: Valid Inferences from Connected and Disjunctive Conditionals

1. The place of exceptive syllogism

The previous lesson treated conjunctive syllogisms: their conclusion follows from the connection between their premises without being expressly stated as a whole in either one.

An exceptive syllogism begins with a conditional relationship and then asserts or denies one of its parts in another premise. A conclusion follows concerning the other part.

For example: “If it rains, the ground will become wet”; “It is raining”; therefore, “The ground will become wet.” The key is not the wording but the relation and the valid use of the second premise.

2. The original definition

وهو مركب من مقدمتين أي قضيتين: إحداهما شرطية والأخرى حملية
It consists of two premises, that is, two propositions: one conditional and the other categorical.

The first premise states a connected or disjunctive relationship. The second asserts or denies one side of that relationship.

Thus an exceptive syllogism is not just a sentence of the form “if ... then ...”. The second premise is needed for a specific conclusion to follow.

3. What “exception” means here

ويتخلل بينهما كلمة الاستثناء أعني إلا وأخواتها — ومن ثم يسمى استثنائياً
An exceptive expression, namely “except” or a corresponding particle, comes between them; hence its name.

In worked arguments, a term such as “but” introduces the proposition being affirmed or denied. It need not be negative: “But the sun has risen” can also supply the exceptive premise.

The technical label must not be confused with the everyday act of merely excluding an item or with negation in every case.

4. Identify the two premises

Consider this argument:
Conditional premise: If the sun has risen, daytime is present.
Exceptive premise: But the sun has risen.
Conclusion: Therefore daytime is present.

The first premise relates antecedent and consequent. The second affirms the antecedent, so the consequent follows.

Crucially, the conclusion does not merely repeat the second premise; the conditional connection allows a different claim to be established.

5. Antecedent and consequent in a connected conditional

In the conditional “If A, then B”, A is the antecedent and B the consequent. The relationship says that if the conditional is true and A occurs, B must follow.

Antecedent: what is supposed, for example sunrise.
Consequent: what follows, for example the presence of daytime.

Knowing that “if A, then B” is true does not by itself establish that A or B has occurred. A separate exceptive premise is needed.

6. First valid rule: affirm the antecedent

فإن كانت الشرطية متصلة فاستثناء عين المقدم ينتج عين التالي
If the conditional is connected, asserting the very antecedent yields the very consequent.

Form:
If A, then B.
But A.
Therefore B.

The conditional states that B follows from A, while the second premise establishes A. Therefore B follows. The inference is valid because of its form, not because of any particular wording.

7. The sun-and-daylight example

كلما كانت الشمس طالعة كان النهار موجودا
Whenever the sun has risen, daytime is present.

لكن الشمس طالعة
But the sun has risen.

Together these yield “فالنهار موجود”, meaning “therefore daytime is present.”

The sun’s rising affirms the antecedent, and daytime affirms the consequent. Once the conditional and the second premise are accepted, the conclusion follows.

8. The condition behind affirming the antecedent

Suppose we reason: “If this figure is a square, it has four sides.” We then establish that it is a square. Four-sidedness follows because of the true conditional.

But if the conditional is false — for example, “If something is a stone, it is alive” — asserting its antecedent does not make the consequent factually true.

Validity of an inference and truth of its premises are different questions. A valid exceptive form guarantees that a false conclusion cannot follow from premises that are all true.

9. Second valid rule: deny the consequent

واستثناء نقيض التالي ينتج رفع المقدم
And excepting the negation of the consequent yields denial of the antecedent.

Form:
If A, then B.
But not B.
Therefore not A.

If A were true, the true conditional would require B. Since B is denied, A cannot still be affirmed. This is a proof that proceeds by denying the consequent.

10. The textbook example of denying daylight

لكن النهار ليس بموجود
But daytime is not present.

The previously accepted conditional states that sunrise entails daytime. The exceptive premise denies daytime, so the textbook concludes:
فالشمس ليست بطالعة
Therefore the sun has not risen.

The conclusion does not arise because “sun” and “day” have the same meaning. It follows from the connection asserted by the original conditional.

11. Worked example: numbers and evenness

Suppose: “If a number is divisible by 4, it is even.” We establish: “This number is not even.”

Antecedent: divisibility by 4.
Consequent: being even.
Exceptive premise: not even.

Therefore the number is not divisible by 4. This follows because any number divisible by 4 must be even.

12. Invalid move: affirming the consequent

Invalid argument:
If a number is divisible by 4, it is even.
But this number is even.
Therefore it is divisible by 4.

This does not follow: 6 is even but not divisible by 4. There can be many ways in which the consequent is true.

Thus “If A then B; B; therefore A” is invalid. The conditional presents A as sufficient for B, not as B’s only possible condition.

13. Invalid move: denying the antecedent

Invalid argument:
If a number is divisible by 4, it is even.
But it is not divisible by 4.
Therefore it is not even.

The number 6 refutes this: it is even although it is not divisible by 4.

Therefore denying the antecedent does not imply denying the consequent. The original conditional only licenses the direction from A to B; a converse requires separate grounds.

14. The four moves in connected conditionals

1. Affirm A: If A then B; A; therefore B — valid.
2. Deny B: If A then B; not B; therefore not A — valid.
3. Affirm B: If A then B; B; therefore A — generally invalid.
4. Deny A: If A then B; not A; therefore not B — generally invalid.

The last two conclusions might happen to be true in a particular situation, but they do not follow from the conditional alone. Truth by coincidence must be distinguished from logical entailment.

15. The different structure of disjunction

A disjunctive conditional does not connect A and B as “if A then B.” Instead, it states an either ... or ... relationship. The force of that relationship depends on whether the two alternatives can both be true or both be false.

The book distinguishes three cases:
True disjunction — neither both true nor both false.
Exclusion of conjunction — they cannot both be true, but they can both be false.
Exclusion of emptiness — they cannot both be false, but they can both be true.

These differences determine which inferences are valid.

16. True disjunction: both directions work

وإن كانت منفصلة حقيقية فاستثناء عين أحدهما ينتج نقيض الآخر وبالعكس
If the conditional is a true disjunction, asserting either side yields the negation of the other, and conversely.

For example, an integer is either even or odd. It cannot be both, nor can it be neither.

Thus affirming evenness denies oddness, while denying evenness establishes oddness. The corresponding inferences also work from the other side.

17. True disjunction: affirmation yields denial

Disjunction: This integer is even or odd.
Exceptive premise: But it is even.
Conclusion: Therefore it is not odd.

This works because an integer cannot be both even and odd. Accepting oddness as well would assert incompatible descriptions of the same number.

The inference uses the fact that both cannot be true together. It does not yet need the further claim that both cannot be false.

18. True disjunction: denial yields affirmation

Disjunction: This integer is even or odd.
Exceptive premise: But it is not even.
Conclusion: Therefore it is odd.

This uses the other condition: even and odd exhaust the possibilities for integers, so both cannot be false.

Key distinction: the earlier argument used affirmation to deny the alternative; this one uses denial to establish the alternative. True disjunction licenses both patterns.

19. Excluding conjunction: what is excluded?

وفي مانعة الجمع ينتج القسم الأول دون الثاني
In a disjunction excluding conjunction, the first kind of inference works, but the second does not.

Exclusion of conjunction means the sides cannot both be true. For example: “This shape is a triangle or a square.” It cannot be both in the same respect, but a circle is neither. Here “or” expresses incompatibility; it does not assert that every shape belongs to one of the two classes.

Thus affirming one alternative rules out the other; denying one does not establish the other because a third possibility remains.

20. Excluding conjunction: valid versus invalid

Valid: This shape is a triangle or a square; it is a triangle; therefore it is not a square.

Invalid: This shape is a triangle or a square; it is not a triangle; therefore it is a square.

A circle refutes the second inference: it is neither a triangle nor a square. The relation only excludes their conjunction; it does not eliminate every third possibility.

With the same “or” statement, a different kind of exceptive premise changes what may validly be concluded.

21. Excluding emptiness: what is excluded?

وفي مانعة الخلو القسم الثاني دون الأول
In a disjunction excluding emptiness, the second kind of inference works, but not the first.

Exclusion of emptiness means the alternatives cannot both be false; they may both be true. Example: “This number is at least 5 or at most 5.” At 5, both statements are true; at other values, at least one is true.

Therefore denying either side establishes the other. But affirming one does not rule out the other, since both may hold.

22. Excluding emptiness: valid versus invalid

Valid: The number is at least 5 or at most 5; it is not at least 5; therefore it is at most 5.

Invalid: The number is at least 5 or at most 5; it is at least 5; therefore it is not at most 5.

The number 5 is a counterexample: it is both at least 5 and at most 5.

What this kind of disjunction guarantees is that both cannot be denied. That is why denial of one yields affirmation of the other.

23. Comparing the three disjunctive rules

True disjunction: the alternatives are neither both true nor both false. Affirmation yields denial, and denial yields affirmation.

Exclusion of conjunction: the alternatives cannot both be true. Affirmation yields denial; denial does not necessarily yield affirmation.

Exclusion of emptiness: the alternatives cannot both be false. Denial yields affirmation; affirmation does not necessarily yield denial.

The mere word “or” does not determine the type. One must examine whether both sides can hold together and whether both can fail together.

24. The first and second forms explained

After treating true disjunction, the book’s “first kind” means affirming one side to deny the other. Its “second kind” means denying one side to affirm the other.

The later rules then become clear:
Exclusion of conjunction allows the first kind, not the second.
Exclusion of emptiness allows the second kind, not the first.

Mistaking the “first kind” for the “first alternative” would distort the argument. The division is by mode of inference, not by the position of the alternatives.

25. Common errors in exceptive arguments

Mistake 1: “But” always introduces a negation. In fact, it may introduce an affirmative exceptive premise.

Mistake 2: Affirming the consequent of any connected conditional establishes its antecedent. Not unless the converse relation is separately known.

Mistake 3: Denying one side of every disjunction establishes the other. This fails in an exclusion-of-conjunction case.

Mistake 4: Every “or” statement has exactly one true side. In an exclusion-of-emptiness case, both sides may be true.

26. Worked examples: choose the valid pattern

Example 1: If it is a square it has four sides; it is a square; therefore it has four sides. Valid affirmation of the antecedent.

Example 2: If it is a square it has four sides; it has four sides; therefore it is a square. Invalid affirmation of the consequent; a non-square rectangle is a counterexample.

Example 3: It is a triangle or square; not a triangle; therefore a square. Not licensed by exclusion of conjunction alone; it might be a circle.

Example 4: The number is at least 5 or at most 5; not at most 5; therefore at least 5. Valid denial-to-affirmation in exclusion of emptiness.

27. A method for checking conclusions

First identify whether a new conditional is connected or disjunctive. In a connected conditional, identify antecedent and consequent; in a disjunction, check the possibility of both being true or both being false.

Then ask which side the exceptive premise affirms or denies and apply the appropriate rule.

If unsure, search for a counterexample: can both premises be true while the proposed conclusion is false? If yes, the conclusion does not logically follow. This tests understanding rather than rote memorization.

28. Summary and transition

An exceptive syllogism uses two premises: a conditional proposition and a categorical premise affirming or denying one of its sides.

Connected conditional: affirming the antecedent establishes the consequent; denying the consequent denies the antecedent. The two converse patterns are not generally productive.

True disjunction: both affirmation-to-denial and denial-to-affirmation are valid.
Exclusion of conjunction: only affirmation-to-denial is valid.
Exclusion of emptiness: only denial-to-affirmation is valid.

The book next turns to induction (istiqrāʾ), where a general judgment is proposed from an examination of particular cases.