Anumana in Indian Philosophy

by Sangita Chakravarty | 2016 | 48,195 words

This page relates ‘Fallacy (Hetvabhasa)’ of the study on the concept of Anumana (inference) in the Vedic schools of Indian Philosophy. Anumana usually represents the most authentic means of valid knowledge. This paper discusses the traditional philosophical systems such as Nyaya, Vaisheshika, Samkhya, Yoga, Mimamsa and Vedanta.

In Indian logic, fallacy is called hetvābhāsa. Hetvābhāsa is defined as something which seems to be a true hetu but which is wrong and which prevents us from arriving at an anumāna. Literally speaking, hetvābhāsa means a fallacious reason which looks like a hetu but is not hetu (hetuvadābhāsate iti...) . In the Tarkasaṃgrahadipīkā, hetvābhāsa is defined as that which is the subject of a right knowledge that prevents a judgment.[1]

S.C. Chatterjee rightly holds,

“The Nyāya account of the fallacies of inference is accordingly limited to those of its members or constituent propositions and these have been finally reduced to those of hetu or reason”.[2]

For the purpose of proof of an inference (anumāna), Gautama mentions five constituents of inference (anumāna), viz.,

  1. pratijñā,
  2. hetu,
  3. udāharaṇa,
  4. upanaya and
  5. nigamana.

The validity of an inference depends on the validity of these five constituent parts of it. If there is anything wrong with any of its members, the syllogism as a whole becomes fallacious. The fallacies of inference (anumāna) ultimately arise out of the fallacious reason (hetvābhāsa). There are no formal fallacies in the Nyāya system, which is not concerned with formal truth. The Naiyāyikas bring the fallacies of inference (anumāna) under the fallacies of the reason.

Different kinds of Hetvābhāsa:

In the Nyāya system Gautama classified all the inferential fallacies under five heads,[3] viz., (i) savyabhicāra, (ii) viruddha, (iii) prakaraṇasama, (iv) sādhyasama, and (v) kālātīta or bādhita.

The fallacies of reason have been variously named and classified by various logicians, e.g.,


  1. savyabhicāra,
  2. viruddha,
  3. prakaraṇasama,
  4. sādhyasama,
  5. kālātīta (bādhita).


  1. savyabhicāra,
  2. viruddha,
  3. satpratipakṣa,
  4. asiddha,
  5. bādhita.


  1. asiddha,
  2. viruddha,
  3. anaikāntika,
  4. kālātyayāpadiṣṭa,
  5. prakaraṇasama.


  1. aprasiddha,
  2. viruddha,
  3. saṃdigdha.

Praśastapāda: adds anadhyavasita in the list of Kaṇāda, but later on includes it under the sub types of asiddha


  1. asiddha,
  2. viruddha,
  3. anaikāntika,
  4. prakaraṇasama,
  5. kālātyayāpadiṣṭa.


  1. savyabhicāra,
  2. viruddha,
  3. satpratipakṣa,
  4. asiddha,
  5. avādhita.[4]


  1. anaikāntika,
  2. viruddha,
  3. asiddha,
  4. pratipakṣita,
  5. kālātyayāpadiṣṭa.

Jayanta also follows Gautama in naming the fallacies of reason but he surpasses almost all his predecessors in giving the detailed account of all the five types of fallacies of reason. Except for a slight difference in nomenclature, the Naiyāyikas and the Vaiśeṣikas are unanimous in holding that the fallacies of reason are five-fold. Now we shall discuss them in the following heads:

(1) Savyabhicāra:

In the Nyāya system, the first kind of inferential fallacy is called savyabhicāra or anaikāntika.[5] There are two views about this fallacy of reason. Gautama calls it savyabhicāra or anaikāntika, while Kaṇāda calls it saṃdigdha.[6] Saṃdigdha is the doubtful or inconclusive reason.

He gives two examples, e.g.,

  1. “This is a horse, because it has horns.” This is an example of the former type of fallacy.
  2. “This is a bull, because it has horns.” This is an example of the second type of fallacy.

Praśastapāda interprets the Vaiśeṣikasūtra as mentioning three fallacies, contradictory, unproven and inconclusive reason. He equates aprasiddha with viruddha, asan with asiddha and saṃdigdha with anaikāntika (savyabhicāra).[7]

Gautama defines an inconclusive reason (savyabhicāra) as one which has variable concomitance with the predicate.[8] The savyabhicāra is that reason which leads to more conclusions than one. In savyabhicāra, the hetu is found to lead to no single conclusion but to different opposite conclusions. The savyabhicāra hetu is not uniformly concomitant with the major term. It is related to both the existence and the non-existence of the major term. Therefore, it is called anaikāntika, or an irregular concomitant of the sādhya.

It is of three kinds, viz.,

  1. sādhāraṇa,
  2. asādhāraṇa,
  3. anupasaṃhārī.[9]

The sādhāraṇa or the ordinary fallacy of the irregular middle occurs when the middle term is in some cases related to the major and in the other cases related to the absence of the major.[10] In the Tarkasaṃgraha it is stated that a sādhāraṇa hetu is that where the reason or hetu may be present, when the property to be proved is absent. Here, the middle term is too wide.

For example:–

All knowable objects are fiery;
The hill is knowable;
So, the hill is fiery.

The second form of the savyabhicāra is known as asādhāraṇa or the extraordinary. It is just the opposite of sādhāraṇa, being found neither in sapakṣa nor in vipakṣa. [11] In it, the middle term is too narrow. It is related neither to things in which the major exists nor to those in which it does not exist.[12] Here, the middle term is only present in the minor term.

For example-

Sound is eternal;

Because there is ‘soundness’ or śabdatva. The nature of sound is a peculiar characteristic of the sound. Here, the middle term ‘soundness’ is related only to the minor term ‘sound’. It is found neither in eternal objects like the ‘soul’ nor in other non-eternal things, like ‘a jar.’

The third form of the savyabhicāra is called anupasaṃhārī, or the indefinite.[13] In the anupasaṃhārī, the middle term is non-exclusive and the minor term is all inclusive. These inclusive reasons are defined by Viśvanātha.[14] Here, in the anupasaṃhārī, the middle term is related to a minor term that stands not for any definite individual, but indefinitely for all objects.

For example:

‘All objects are eternal’,
Because they are knowable.

Here, the validity of this inference (anumāna) depends on the validity of the major premise, viz., ‘all knowable objects are eternal’. But the validity of the major premise cannot be proved since beyond all objects we have no instances of the concomitance between the knowable and the eternal. According to Uddyotakara there are sixteen sub-kinds of inconclusive reasons.[15] In the Nyāya system, Jayanta’s view is identical with that of Vātsyāyana, regarding the etymology of the term anaikāntika. According to Jayanta, if a reason does not belong to similar instances, it is faulty and is to be called anaikāntika. [16] In the Tarkabhāṣā, Keśavamiśra opines that savyabhicāra is of two kinds, viz., sādhāraṇa and asādhāraṇa. The common strayer or sādhāraṇa hetu is that which exists in all the three, i.e., pakṣa, sapakṣa and vipakṣa. [17] On the otherhand, the peculiar strayer or asādhāraṇa hetu is that which is absent from both sapakṣa and vipakṣa but exists only in pakṣa. [18]

(2) Viruddha:

The second kind of inferential fallacy is called viruddha. In the Nyāyasūtra[19] and Nyāyabhāṣya, it is stated that the contradictory is the reason which opposes what is be established. For example, ‘a pot is produced, because it is eternal’. Here, the reason is contradictory. Because that which is eternal is never produced. Kaṇāda refers it as aprasiddha. [20] The unsubstantial is called fallacious arguments. According to Uddyotakara, it is called viruddha since it contradicts an admitted truth.[21] For example, ‘sound is eternal’, ‘because it is produced.’ Here the reason, ‘producedness’ contradicts the proposition, ‘sound is eternal’ instead of proving it. Vātsyāyana states that the fallacy of the viruddha consists in the opposition of one doctrine to a previously accepted doctrine.[22] Thus, it is a contradiction between the different parts of a system. In the Tarkasaṃgraha, Annambhaṭṭa opines, that is called viruddha or the contradictory middle, which is pervaded by the absence of the major term.[23] The result is that such a middle term instead of proving the existence of the major in the minor term, which is intended by it, proves its non-existence there in. In the Tarkabhāṣā it is stated that the contradictory reason is that which has invariable concomitance with the negation of the probandum, e.g., sound is eternal, because it is a product.[24] From the definition it is clear that a viruddha hetu can exist only in vipakṣa and so lacks the characteristic vipakṣāsattva. According to Gaṅgeśa, the viruddha hetu proves the non-existence of the predicate, though it is advanced to prove its existence. Jayanta is quite aware of the difference between the interpretations offered by Vātsyāyana and Uddyotakara. In the above discussion, it is observed that there is a slight difference between the fallacies of the savyabhicāra and the viruddha. The difference is that while in the former, the middle term is universally related neither to the existence of the major nor to its non-existence. In the later, the middle term is universally related to the non-existence of the major term. As a whole it means that the savyabhicāra hetu fails to prove the conclusion whereas the viruddha hetu proves the contradictory proposition.

(3) Prakaraṇasama or Satpratipakṣa:

The third kind of inferential fallacy is known as prakaraṇasama or satpratipakṣa. Gautama, in his Nyāyasūtra, states that here the middle term is contradicted by another middle term.[25] Vātsyāyana interprets it as the oscillation of mind between two contradictory characters of an object brought about by mutually opposed and equally strong arguments and counter arguments[26] . It means that when a middle term oscillates between two opposite views we have a case of the prakaraṇasama middle. In the Tarkasaṃgraha, Annambhaṭṭa states that when the reason is counterbalanced by another reason which proves the negation of what is to be proved then it is called the fallacy of equalization or satpratipakṣa. [27] Gaṅgeśa thinks that satpratipakṣa is a temporary flaw in an inference (anumāna) which remains till a doubt about one of the reasons is removed.[28] Keśavamiśra refers that the prakaraṇasama is also called satpratipakṣa or that which is opposed by another reason which proves the existence of the opposite of the proposed probandum.[29] Jayanta refers to asatpratipakṣatva as one of the five characteristics of a valid reason. He maintains that if a reason is faced with the existence of an equally strong counter reason (pratipakṣa), it is fallacious since the opposite reasons counteract with each other and fail to establish the conclusion. He distinguishes a counterbalanced reason (prakaraṇasama) from an inclusive reason (anaikāntika). In the former, two different characters of the minor term are taken as the middle terms leading to opposite conclusions. In the later one, the same character of the minor term is taken as a middle term that may lead to opposite conclusions. Again, it is also different from the viruddha hetu or contradictory middle. The former is not known to exist or not to exist in similar instances; nor is it known to exist in dissimilar instances, while the later is known to exist in dissimilar instances.[30]

(4) Sādhyasama or Asiddha:

The fourth kind of inferential fallacy is called sādhyasama or the asiddha. Gautama defines an unproven reason (sādhyasama) as one that requires to be proved like the predicate.[31] An asiddha hetu is one that has doṣa called asiddhi and asiddhi is the absence of siddhi which consists in the true cognition of the sādhyavyāpyahetu as a property of pakṣa; or briefly speaking, asiddhi is the non-production of parāmarśa. The word sādhyasama means a middle term which is similar to the sādhya or the major term. Vātsyāyana regards it as not different in logical validity from the predicate in as much as it requires to be proved like it. His illustration of this fallacy runs as ‘shadow is a substance, because it possesses motion.’ Here, unless it is really proved that shadow possesses motion, it cannot be accepted as the reason for the proposition that shadow is a substance.[32] Uddyotakara, however, mentions it as asiddha and sub-divides it into three kinds; viz., prajñāpanīyadharmasamāna, āśrayāsiddha and anyathāsiddha. [33] Jayanta refers to this fallacy by both of the names, i.e., sādhyasama and therefore fallacious.[34]

In the Vaiśeṣika system, Praśastapāda refers four sub-types of asiddha. These are—

  1. ubhayāsiddha [ubhayāsiddhaḥ], 
  2. anyatarāsiddha [anyatarāsiddhaḥ], 
  3. tadbhāvāsiddha [tadbhāvāsiddhaḥ] and
  4. anumeyāsiddha [anumeyāsiddhaḥ].[35]

The main forms of the fallacy of asiddha, is of three kinds, viz.,

  1. āśrayāsiddha,
  2. svarūpāsiddha and
  3. vyāpyatvāsiddha.[36]

In the āśrayāsiddha, the minor term is the locus of the middle term. If the minor term is unreal and fictitious, the middle term cannot be related to it; consequently, in such a case, the minor premise in which the middle term is related to a fictitious minor term becomes false, e.g., “a skylotus is fragrant, because it is a lotus, like a lotus in the pond”. Here, skylotus is the subject and that is totally non-existent.[37] The middle term having no locus standi, we have a fallacy of the āśrayāsiddha or the baseless middle. From the example āśrayāsiddha can also be defined as- pakṣe pakṣatāvacchedakābhāvaḥ or simply pakṣatābhāvaḥ. [38]

The svarūpāsiddha is a middle term which cannot be proved to be real in relation to the minor term. It is a middle term which is not found in the minor term.[39] In it, the hetu itself is asiddha. A reason must exist in the subject. If it does not exist, it can not afford to provide the basis of inference. For example–“sound is eternal, because it is visible, like a pot.” Here we find that the visibility is not present in the subject and its assumption in sound is wrong and is not justified by facts. This differs from āśrayāsiddha. In the svarūpāsiddha, the locus is either false or not proper. It may be defined as - pakṣe hetvābhāvaḥ.

The svarūpāsiddha consists in many sub-types, such as—

  1. śuddhāsiddha,
  2. bhāgāsiddha,
  3. viśeṣaṇāsiddha and
  4. viśeṣyāsiddha.

The general characteristic of these sub-varieties is that in all these, the reason is non-existent in the minor term.

The vyāpyatvāsiddha is a middle term whose concomitance (vyāpti) with the major term cannot be proved.[40] A valid inference requires that the middle term must be concomitant with the major term. If this condition is not fulfilled, the inference becomes invalid.

The fallacy of vyāpyatvāsiddha may be of two types—

(i) It may be on account of the non-concomitance of the middle term with the major term. It means the first is caused by the absence of any evidence to grasp the vyāpti.

For example–

all reals are momentary,
sound is real,
therefore, sound is momentary.

There is no evidence which enables one to ascertain the vyāpti, yat sat tat kṣaṇikam. Here, the major premise is false. Because there is no universal relation between the ‘real’ and ‘momentary’.

(ii) Or, it may be on account of the presence of a condition (upādhi). For example–“the hill is smoky, because it is fiery.” Here, the relation of the middle term (fire) to the major term (smoke) is conditional since a fire is smoky if there is a wet fuel.[41] This fallacy of the conditional middle is technically called vyāpyatvāsiddha. [42]

(5) Kālātīta or Bādhita:

The fifth kind of inferential fallacy is called kālātīta or bādhita hetu. The kālātīta literally means a middle term which is vitiated by the lapse of time.[43] Vātsyāyana interprets it in this manner- “sound is durable, because it is manifested by conjunction, like colour.” Here, the argument is fallacious, since in the case of colour the manifestation takes place simultaneously with the contact between light and the coloured thing. But the manifestation of sound is separated by an interval of time from the contact between two objects. So, here, the reason is not congruous with the instance and is not capable of establishing the predicate.[44] Uddyotakara gives the same meaning of kālātyayapadiṣṭa, as Vātsyāyana does.[45] But Vācaspati takes it in the sense of a contradicted reason (bādhita). It is contradicted by perception, inference and Vedic testimony. According to the Tarkasaṃgraha, a bādhita hetu is that where the negation of what is to be proved is established without doubt by another proof.[46] In the Tarkabhāṣā, Keśavamiśra opines that the term kālātyayāpadiṣṭa means a hetu put forward after the lapse of proper time; i.e., after the negation of the probandum has been ascertained by a more trust-worthy proof. It is also called bādhita. [47] Gaṅgeśa clearly states bādhita in the list of fallacious reason. According to Jayanta, in this fallacy the reason refers more than one event which succeed one another in time. He, however, incorporates the absence of uncontradictoriness as a characteristic of kālātyayāpadiṣṭa and thereby seems to hold bādhita as another name of kālātīta. While the kālātīta stands for a middle term vitiated by a limitation in time, the bādhita means a middle term which is contradicted by some other source of knowledge. The fallacy of bādhita is different from the fallacy of satpratipakṣa. In the former, an inference is contradicted by a non-inferential source of knowledge, while in the later one inference is contradicted by another inference.

Footnotes and references:


anumitipratibandhaka yathārthajnānavisayatvam hetvābhāsatvam, Tarkasaṃgrahadīpikā on Tarkasaṃgraha 46


The Nyāya Theory of Knowledge, p. 218


savyabhicāraviruddhaprakaraṇasamasādhyasamakālātītā hetvābhāsaḥ. Nyāyasūtra, i. 2. 4


savyabhicāraviruddhasatpratipakṣāsiddhavādhitāḥ pañca hetvābhāsaḥ. Tarkasaṃgraha, p. 33


anaikāntikaḥ savyabhicāraḥ. Nyāyasūtra, i. 2. 5


aprasiddho’napadeśo’san saṃdigdhaścānapadeśaḥ, yasmād viṣāni tasmād aśvaḥ; yasmād viṣāni, tasmād gauriti cānaikāntikasyodāharaṇam. Vaiśeṣikasūtra, iii. 1. 15-17


viruddhāsiddhasandigdham aliṅgam kāśyapo’bravit. Praśastapāda Bhāṣya, p. 200


anaikāntikaḥ savyabhicāraḥ. Nyāyasūtra, i.ii. 4


sa trividhaḥ, sādhāraṇāsadhāraṇānupasaṃharībhedāt. Tarkasaṃgraha, p. 33 Cf: ādyaḥ sādhāraṇastu syāt syadasādhāraṇaparaḥ / tathaivānupasṃhārī tridhā anaikāntiko bhavet // Bhāṣāpariccheda, 72, p. 388


sādhāraṇaḥ sādhyavat tadanyavṛttiḥ. Nyāyasūtravṛtti, 1. 2. 5


sarvasapakṣavipakṣavyāvṛtto’sādhāraṇaḥ. Tarkasaṃgraha,46


asādhāraṇaḥ sapakṣavipakṣa vyāvṛttaḥ. Nyāyasūtravṛtti, 1. 2. 5


anvayavyatirekadṛṣṭāntarahito’nupasaṃhārī. Tarkasaṃgraha, p. 56


sādhāraṇaḥ sādhyavadanyavṛttiḥ; anye tu sapakṣavṛttirasādhāraṇaḥ; anupasaṃhārī cātyantābhāvāpratiyogī sādhyakādi. Siddhāntamuktāvalī, pp. 388-389 on kārikā 72


Vide Bijalwan, C.D., Indian Theory of Knowledge, p. 177




tatra pakṣasapakṣavipakṣavṛttiḥ sādhāraṇaḥ. Tarkabhāṣā, 55


yasapakṣādvipakṣāccha vyāvṛttaḥ pakṣe eva vartate so’sādhāraṇaḥ. Ibid.


siddhāntamabhyupetya tadvirodhī viruddha. Nyāyasūtra, i.ii. 6


aprasiddho’napadeśaḥ. Vaiśeṣikasūtra, iii. 1. 15


pratijñāhetvorvā virodhaḥ. Nyāyavārttika, 1. 2. 6


Nyāyabhāṣya, 1. 2. 6


sādhyābhāvavyāpto heturviruddhaḥ. Tarkasaṃgraha, p. 57


sādhyaviparyayavyāpto heturviruddhaḥ, sa yathā śabdo nityaḥ kṛtakatvāt gagaṇavat. Tarkabhāṣā, 54


yasmāt prakaraṇacintā sa nirṇayārtham apadiṣṭaḥ prakaraṇasamaḥ. Nyāyasūtra, i. 2. 7


Nyāyabhāṣya, i. 2. 7


sādhyābhāvasādhakaṃ hetvāntaraṃ yasya vidyate sa satpratipakṣaḥ. Tarkasaṃgraha, 46


prakaraṇasamastu sa eva yasya sādhyaviparītasādhakaṃ hetvantaraṃ vidyate. Tarkabhāṣā, 56


Nyāyamañjarī, pp. 602-603; Nyāyasūtravṛtti, 1. 2. 6


sādhyāviśiṣṭaḥ sādhyatvāt sādhyasamaḥ. Nyāyasūtra, i. 2. 8


Nyāyabhāṣya, I.II. 8


Vide.Sinha, J,N., Indian Philosophy, (vol-1) p. 531


Nyāyamañjarī, II-162


tatrāsiddhaścaturvidhaḥ ubhayāsiddhaḥ,anyatarāsiddhaḥ,tadbhāvāsiddhaḥ,anumeyāsiddhaśceti. Praśastapāda Bhāṣya, p. 577


tatra asiddhastṛvidhaḥ, āśryāsiddhaḥ svarūpāsiddho vyapyatvāsiddha iti. Tarkabhāṣā, 53


gagaṇāravindaṃ surabhi, aravindatvāt, sorojāravindavat. atra gagaṇāravindamāśrayaḥ sa tu nāstyeva. Ibid.


Ibid., p. 98


svarūpāsiddho yathā, śabdo’nityaḥ cākṣuṣatvāt, ghatavat. Ibid.


pakṣe vyāpyatāvacchedakābhāvaḥ. Ibid.




kālātyayāpadiṣṭaḥ kālātītaḥ. Nyāyasūtra, i. 2. 9


kālātyayena yukto yasyārthasyaikadeśo’padiśyamānasya sa kālātyayāpadiṣṭaḥ kālātīta iti uchyate…..evamudāharaṇasādharmasyābhāvādasādhanamayaṃ heturhetvābhāsa iti. Nyāyabhāṣya, i. 2. 9


Nyāyavārttika, i. 2. 50


yasya sādhyabhāvaḥ pramāṇāntarena niścitaḥ sa bādhitaḥ. Tarkasaṃgraha, 46


pakṣe pramāṇāntarāvadhṛtasādhyābhāvoheturbādhitaviṣayaḥ kālātyayāpadiṣṭa iti cochyate. Tarkabhāṣā, 57

Like what you read? Consider supporting this website: