sanskrit
stringlengths
2
508
english
stringlengths
2
924
sāmānādhikaraṇyam / kvopayujyata ity arthaḥ /
[The term] sāmānādhikaraṇyam [means] co-ordination. The meaning is "where would [it] be useful?"
sato 'pyajñātasya śābde vyavahāre 'naṅgatvāt //
Because even if something exists, if it is not cognized, it cannot form part of verbal usage.
athāpi syāt yadi vyāvṛttimātraṃ śabdenocyeta tadāyaṃ doṣaḥ syāt yāvatā vyāvṛttimadvastu vācyam ato vyāvṛttidvayopādhikayoḥ śabdayor ekasminnapohavati vastuni vṛtteḥ sāmānādhikaraṇyam astyevetyata āha athetyādi / athāpyapohavad vastu vācyamityabhidhīyate /
[One] might argue thus: "If mere exclusion were what is expressed by the word, then this fault would exist. However, what is to be expressed is an entity possessing exclusion. Therefore, since two words qualified by two exclusions can occur in reference to a single entity possessing apoha, co-ordination definitely exists."
tatrāpīti apohavati vastuni /
"There also" [means] "in the entity possessing apoha."
vācyatvenāṅgīkriyamāṇe 'nīlādibhedānām utpalādīnāṃ nīlādiśabdair vyāptirākṣepo durlabhaḥ kiṃ kāraṇam/ paratantratvānnīlādiśabdasya, sa hi vyāvṛttyupasarjanaṃ tadvantamarthamāha na sākṣāt /
When an entity possessing apoha is accepted as what is to be expressed, the indication of the concomitance of various non-blue things like lotuses by words like "blue" would be difficult to obtain. Why? Because words like "blue" are dependent [on something else], for they express an object as subordinate to exclusion, not directly.
tataś ca sākṣādanabhidhānāt tadgatabhedākṣepo nāsti yathā madhuraśabdena śuklādeḥ /
And thus, due to the absence of direct denotation, there is no indication of the differences inherent in it, just as the word "sweet" [does not indicate] white and so forth.
yadyapi vastusthityā śuklādīnāṃ madhurādibhedatvam asti / tathāpi śabdasya sākṣādabhihitārthagatasyaiva bhedasyākṣepasāmarthyaṃ na
Although in reality white and other [qualities] are varieties of non-sweet and so forth, nevertheless the word has the power to indicate only the varieties of what it directly denotes, not [the varieties] of what is denoted through something else.
tataś ca nīlādiśabdena tadgatabhedānākṣepād utpalādīnām atadbhedatvaṃ syāt atadbhedatve ca sāmānādhikaraṇyaṃ na prāpnoti /
Therefore, since words like "blue" do not indicate the varieties inherent in it, lotuses and such would not be its varieties, and when they are not its varieties, co-ordination would not be possible.
tena jātiman mātrapakṣe yo doṣa ukto bhavatā tadvato na bādhakaḥ śabdo 'svatantratvād iti sa vyāvṛttimanmātrapakṣe 'pi tulya iti darśitaṃ bhavati /
Thus, the objection that you raised against the view that [what is denoted is] merely something possessing a universal—namely that "the word cannot denote that which possesses [the universal] because it is not independent"—is shown to apply equally to the view that [what is denoted is] merely something possessing exclusion.
tathā hi jātimanmātre śabdārthe sacchabdo jātisvarūpopasarjanaṃ dravyamāha na sākṣād iti tadgataghaṭādibhedānākṣepādatadbhedatve sāmānādhikaraṇyābhāvaprasaṅga uktaḥ
When the word-meaning consists only of [something] possessing a universal, the word "sat" expresses substance with the form of the universal as its subordinate factor, [and] not directly; because it does not indicate the varieties like jar etc. inherent in it, and when these varieties are not indicated, there would be no possibility of co-reference, as has been stated.
sa vyāvṛttimanmātrapakṣe 'pi samānaḥ tatrāpi hi sacchabdo vyāvṛttyupasarjanaṃ dravyam āha na sākṣād iti tadgatabhedānākṣepo 'trāpi samāna eva ko hyatra viśeṣo jātir vyāvṛttir jātimānvyāvṛttimān iti
This [defect] applies equally to the view that [the meaning consists of something] possessing only exclusion, for there too the word "sat" expresses substance with exclusion as its subordinate factor, [and] not directly; the non-indication of varieties inherent in it is exactly the same here too - for what is the difference here between universal and exclusion, [or between] universal-possessor and exclusion-possessor?
doṣāntaram āha liṅgetyādi
[The text] states another defect beginning with "liṅga" etc.
liṅgasaṅkhyādisambandho na cāpohasya vidyate / vyakteścāvyapadeśyatvāt tadvāreṇāpi nāstyasau
There exists no connection between apoha and gender, number etc.; and since the particular is inexpressible, there is no [connection] through that either.
liṅgāni strītvapuṃstvanapuṃsakāni saṅkhyā ekatvadvitvabahutvāni ādigrahaṇena kriyākālādisambandhaparigrahaḥ
"Gender" [means] feminine, masculine and neuter; "number" [means] singular, dual and plural; the term "etc." includes the connection with action, time and so forth.
ebhiḥ sambandho nāstyapekṣasya avastutvāt eṣāṃ ca vastudharmatvāt
There is no connection between these [gender etc.] and apoha, because it [apoha] is not a real entity, while these [gender etc.] are properties of real entities.
na liṅgādiviviktaḥ padārthaḥ śakyaḥ śabdenābhidhātum
A word-meaning devoid of gender etc. cannot be expressed by a word.
athāpi syād vyāvṛttyādhārabhūtāyā vyakter vastutvālliṅgādisambandho 'sti etadvāreṇāpohasyāpi vyavasthāpyata ity āha vyakteścetyādi
If one might argue that "because the particular which serves as the substratum of exclusion is a real entity, it has connection with gender etc., and through this [connection] apoha too can be established," [the text] states "and since the particular..." etc.
sā hi vyaktir nirvikalpatvālliṅgasaṅkhyādisambandhena vyapadeṣṭuṃ na pāryate tataḥ kathamapohe tadvāreṇa tadvyavasthā siddhyet
For that particular, being indeterminate, cannot be designated as connected with gender, number etc.; therefore how could their establishment in apoha be accomplished through it?
avyāpitvaṃ cāpohaśabdārthavyavasthayor darśayann āha ākhyāteṣvityādi
Showing the non-pervasiveness of the establishments of apoha and word-meaning, [the text] states "in the case of finite verbs" etc.
ākhyāteṣu tiṅanteṣu pacati gacchatītyevam ādiṣu kriyāpradhāneṣu śabdeṣunānyavyavacchedo gamyate kasmād ity āha na paryudāsarūpaṃ hītyādi
In the case of finite verbs, words with verbal endings like "pacati" ["cooks"], "gacchati" ["goes"] and so forth, where action is the primary [meaning], the exclusion of others is not understood; [if asked] "why?", [the text] states "because there is no form of exclusion" etc.
yathā subanteṣu ghaṭādiśabdeṣu niṣpannarūpamaghaṭādikaṃ paryudāsarūpaṃ niṣedhyamasti na tathā pacatītyādiṣu pratiyogino niṣpannasya kasyacid apratīteḥ
While in the case of declined nominal stems like "ghaṭa-" etc., there exists a well-established entity to be negated in the form of a paryudāsa [namely] the non-jar etc., this is not so in the case of [verbal forms] like "pacati" etc., since no established counter-entity is ever apprehended.
na neti hyucyamāne 'pi niṣedhasya niṣedhanam
Even when "na na" [double negation] is being uttered, [there would be only] a negation of negation;
pacatītyaniṣiddhaṃ tu svarūpeṇaiva tiṣṭhati
but [the verbal form] "pacati" remains unnegated in its own form.
sādhyatvapratyayaścātra tathā bhūtādirūpaṇam
And here [in verbs] there is both the notion of being-in-process-of-accomplishment and the expression of past time etc.;
niṣpannatvād apohasya nirnimittaṃ prasajyate
[and] because apoha is [already] established, [these verbal properties] would [absurdly] become groundless.
na ca pacatītyevam ucyamāne 'pi prasajyapratiṣedhasya niṣedhanamevoktaṃ syāt
Even when it is expressed thus as "na na pacati", what would be expressed is only the negation of a prasajya-pratiṣedha [complete negation].
tataś ca ko doṣa ity āha pacatītyaniṣiddham ityādi
[To the question] "What then is the fault in that?", [the author] states "pacati remains unnegated" etc.
pratiṣedhadvayasya vidher viṣayatvāt pacatītyetat svarūpeṇaiva vidhyātmanaiva tiṣṭhatīti vidhir eva śabdārthaḥ syāt
Since a double negation has affirmation as its object, this [form] "pacati" remains in its own form, that is, in its affirmative nature, [and] thus the meaning of the word would be affirmative only.
kiñca atra pacatītyādau sādhyatvaṃ pratīyate
Moreover, in this [verbal form] "pacati" etc., being-in-process-of-accomplishment is understood.
yasyāḥ kriyāyāḥ kecid avayavā niṣpannāḥ kecid aniṣpannāḥ sā pūrvāparībhūtāvayavā kriyā sādhyatvapratyayaviṣayā
That action of which some parts are accomplished [and] some unaccomplished, [that] action whose parts occur in sequential order, is the object of the notion of being-in-process-of-accomplishment.
tathābhūd bhaviṣyatītyādau bhūtādikālaviśeṣapratītir asti /
Similarly, in [forms] such as "abhūt" and "bhaviṣyati", there exists a cognition of specific temporal distinctions of past and [future] etc.
nacāpohasya sādhyatvādisambhavaḥ niṣpannatvād abhāvaikarasatvena /
And there can be no possibility of [the notion of] being-in-the-process-of-accomplishment etc. for apoha, because [it is] already accomplished, consisting solely of negation.
tasmād apohaśabdārthapakṣe sādhyatvapratyayo bhūtādipratyayaś ca nirnimittaḥ prāpnotīti pratītibādhā /
Therefore, on the view that apoha is the meaning of words, both the notion of being-in-the-process-of-accomplishment and the notion of past etc. would be without cause - hence [this] contradicts [actual] cognition.
bhūtādirūpaṇam iti /
[This refers to] the conceptualization of past etc.
vidhyādāvartharāśau ca nānyāpohanirūpaṇam / naṭaścāpi naṭā yuktāvapohaḥ kīdṛśo bhavet //
In the collection of meanings such as injunction etc., there is no conceptualization of exclusion-of-others; and when a negative is combined with [another] negative, what kind of apoha could there be?
cādīnām api naṭyogo naivāstītyanapohanam /
And since particles like "ca" etc. cannot be combined with negation, [there can be] no exclusion.
vākyārthe 'nyanivṛttiś ca vyapadeṣṭuṃ na śakyate //
And in the meaning of a sentence, the exclusion-of-others cannot be indicated.
ananyāpohaśabdādau vācyaṃ na ca nirūpyate /
And in expressions like "ananyāpoha" etc., no denotation can be ascertained.
prameyajñeyaśabdāder apohyaṃ kuta eva tu //
And from where indeed [could there be] anything excludable in the case of words like "prameya" and "jñeya" etc.?
ādiśabdena nimintraṇāmantraṇādīnāṃ grahaṇam anyāpohasya anyavyavacchedasya na nirūpaṇam upalambho 'sti /
By the word "ādi", invitation, address etc. are included; [and for these] there is no conceptualization [or] apprehension of exclusion-of-others, [i.e.] the differentiation from others.
"na paryudārūpaṃ hi niṣedhyaṃ tatra vidyata" ityetat pūrvoktamevātra kāraṇam iti bhāvaḥ /
The meaning is that the reason for this is exactly what was stated before: "indeed, there exists nothing that could be denied in the form of what is to be excluded."
na na pacati devadatta iti naṭo 'pareṇa naṭā yoge sati kīdṛśo 'poho bhavet / naiva kiñcit /
In [the expression] "na na pacati devadattaḥ" ("Devadatta is not non-cooking"), when one negative is combined with another negative, what kind of apoha could there be? Nothing whatsoever.
pratiṣedhadvaye vidhereva saṃsparśanāt /
Since in [the case of] two negatives, there is always contact with the affirmative alone.
apica cādīnāṃ nipātopasargakarmapravacanīyānāṃ padatvam iṣṭam nacaiṣāṃ naṭā sambandho 'styasattvavacanatvāt /
Moreover, particles like "ca" – that is, nipātas [indeclinables], upasargas [prefixes], and karmapravacanīyas [prepositions] – are accepted as words, yet these cannot have any connection with [the particle] "na" since [such a combination] would not express any [real] existence.
tathā hi yathā ghaṭādiśabdānām aghaṭa ityādau naṭā sambandhe satyarthāntarasya paṭādeḥ parigrahāt tadvyavacchedena naṭā rahitasya ghaṭaśabdasyārtho 'vakalpyate na tathā cādīnāṃ nacetyādi naṭā sambandho 'sti /
For instance, when words like "ghaṭa" [jar] are connected with [the particle] "na" in expressions like "aghaṭa" [non-jar], there is comprehension of something else like "paṭa" [cloth], and thus the meaning of the word "ghaṭa" without [the particle] "na" is understood through the exclusion of that [other thing]; [but] such a connection with "na" does not exist for particles like "ca" in expressions like "na ca."
nacāsabandhyamānasya naṭāpohanaṃ yuktam /
And what is not connected with [the particle] "na" cannot logically be subject to apoha [exclusion].
ataścādiṣvanapohanam /
Therefore, in [the case of] "ca" and similar [particles], there is no apoha.
apohābhāva ity arthaḥ /
The meaning is [that there must be] absence of apoha.
kiñca vākyārthaḥ kalmāṣavarṇavacchavalaikarūpa iṣyate atas tatrānyanivṛttir vyapadeṣṭuṃ na śakyate /
Moreover, the meaning of a sentence is accepted to be of one mixed form, like a variegated color; therefore, the exclusion of others cannot be indicated in that [case].
niṣpannarūpasya pratiyogino 'pratīteḥ /
Because no counter-entity of [such] a completed form is cognized.
yā cātra caitra gāmānayetyādāvacaitrādivyavacchedarūpānyanivṛttiravayavaparigraheṇa varṇyate sā padārtha eva syān na vākyārthaḥ /
And what is described here through the acceptance of parts as the exclusion of others in the form of [excluding] non-Caitra etc. in [sentences] like "Caitra, bring the cow" would be merely the meaning of words, not the meaning of the sentence.
tasyānavayavasyetthaṃ vivektumaśakyatvādityavyāpinī śabdārthavyavasthā /
Since that [sentence-meaning], being without parts, cannot be analyzed in this way, [your] theory of word-meaning is [thus] non-pervasive.
nānyāpoho 'nanyāpoha ityādau śabde vidhirūpādanyadvācyaṃ na nirūpyate nopalabhyata ity arthaḥ
In expressions such as "nānyāpohaḥ ananyāpohaḥ" [and similar ones], nothing other than something of positive form is identified or apprehended as the denotation.
tathā hi nāpoharūpamātraṃ vācyaṃ gamyate pratiṣedhadvayena vidher evāvasayāt
For indeed, what is understood as the denotation is not merely the form of exclusion (apoha), since a double negation results in the ascertainment of something positive.
ādiśabdenānanyavyāvṛttirananyavyavaccheda ityādiparyāyagrahaṇam
By the word "ādi" [in the compound ananyāpohaśabdādau], synonyms such as "ananyavyāvṛtti" and "ananyavyavaccheda" [and others] are included.
nanu ca naṭāścāpi naṭā yuktāvityanenaiva gatatvāt punar uktam etat
[Objection:] "Isn't this merely a repetition, since this point has already been covered by the statement 'naṭāścāpi naṭā yuktau'?"
satyam etat
This is true.
kiṃtvanyāpohaḥ śabdārtha ityevaṃ vādinā svavacanenaiva vidhiriṣṭa iti jñāpanārthaṃ punar ucyate
However, this is repeated to show that the Apohavādin himself, through his own statement "anyāpohaḥ śabdārthaḥ," accepts something positive [as the meaning].
tathā hyanyāpohaśabdasyānanyāpohaśabdārtho vyavacchedyaḥ sa ca vidher nānyo lakṣyate
For the term "anyāpoha" [exclusion of others] excludes what is denoted by the term "ananyāpoha" [non-exclusion of others], and that is observed to be nothing other than something positive.
ye ca prameyajñeyābhidheyādayaḥ śabdās teṣāṃ na kiñcidapohyam asti
And for words such as "prameya" [cognizable], "jñeya" [knowable], "abhidheya" [predicable] and others, there is nothing that can be excluded.
sarvasyaiva prameyatvādisvabhāvatvāt
Because everything has the nature of being cognizable and so forth.
tathā hi yannāma kiñcid vyavacchedyameṣāṃ kalpyate tat sarvaṃ vyavacchedākāreṇālambyamānaṃ jñeyādisvabhāvamevāvatiṣṭhate nahyaviṣayīkṛtaṃ vyavacchettuṃ śakyate
For indeed, whatever might be assumed to be excluded by these [words] would, when grasped in the form of exclusion, still remain of the nature of being knowable etc., since what is not made an object [of cognition] cannot be excluded.
nanu hetumukhe nirdiṣṭam ajñeyaṃ kalpitaṃ kṛtvā tadvyavacchedena jñeye 'numānam iti
[It was] stated in the section on inference that, having assumed something unknowable, [one makes] an inference about the knowable through the exclusion of that [unknowable].
tat katham avyāpitvaṃ śabdārthavyavasthāyā ity āha apohyetyādi
How [can there be] non-pervasiveness of the word-meaning relation? Thus [the author] states [beginning with] "apohya."
jñānākāraniṣedhāc ca nāntarārtho 'bhidhīyate
Since [we have] rejected [the theory of] cognition-forms, no internal object is denoted.
na cāpyapohyatā tasmān nāpohasteṣu siddhyati
And therefore no excludability [exists], [so] apoha is not established regarding these.
evam ityādiśabdānāṃ na cāpohyaṃ nirūpyate
Thus no excludable [object] is determined for words like "evam" [and] so forth.
yadi jñeyam api sarvaṃ jñeyatvenāpohyam asya kalpyate tadā varaṃ vastveva vidhirūpam eva śabdārthatvena vikalpitaṃ bhavedyadadhyavasīyate lokena
If everything knowable is assumed to be excluded qua being knowable, then it would be better to conceptualize the positive entity itself as the word-meaning, which is [what is] determined by people.
evaṃ hyadṛṣṭāropo dṛṣṭārthāpalāpaś ca na kṛtaḥ syād iti bhāvaḥ
For thus there would be neither the superimposition of the unseen nor the denial of the seen - this is the meaning.
ata eva varam ityuktam
Therefore precisely it is said "[it would be] better."
ye tvāhur vikalpapratibimbakam eva sarvaśabdānām arthas tadeva cābhidhīyatevyavacchidyata iti cocyata iti
But those who say that the meaning of all words is only the reflection of conceptual thought, and that this very [reflection] is what is denoted, excluded, and expressed...
tān pratīdamāha jñānākāraniṣedhāccetyādi
Against them, this [verse beginning with] "because of the rejection of cognition-forms" etc. is stated.
nirākārā buddhir ākāravān bāhye 'rthaḥ sa ca bahirdeśasambandho vispaṣṭamupalabhyata ityevam asmābhir jñānākāro niṣiddhaḥ
[We maintain that] cognition is formless [while] the external object has form, and it is clearly perceived as having a connection to external space - thus we have rejected [the theory of] cognition-forms.
tasmād āntarasya buddhyārūḍhasyākārasyāsattvāt tadabhidhāyakatvaṃ śabdānām ayuktam
Therefore, due to the non-existence of an internal form residing in cognition, it is incorrect [to maintain] that words denote that [internal form].
nāpi tasyāntarasyārthasyāpohyatvaṃ yuktam asattvād eva
Nor is the excludability of that internal object possible, precisely because of [its] non-existence.
teṣviti prameyādiśabdeṣu /
[This is said] in regard to those words like "prameya" [cognizable] etc.
ye ca evam ittham ityādayaḥ śabdās teṣām api na kiñcid apohyamupalabhyate, kasyacit pratiyoginaḥ paryudāsarūpasyābhāvāt /
And in the case of such words as "evam" [thus] and "ittham" [in this way] etc., nothing is found that can be excluded, because there exists no counter-entity in the form of what could be excluded.
athāpi syāt naivam ityādi prasajyarūpaṃ niṣedhyaṃ bhavatīti /
If it might be argued that in such [expressions] as "na evam" [not thus] etc., there is something that could potentially be excluded...
naitad astyuktam atra /
This [argument] is not valid here, as [already] stated.
tathā hi atrāpi na naivam iti nirdeśe niṣedhasya niṣedhanam, evam ityaniṣiddhaṃ tu svarūpeṇaiva tiṣṭhatīti saiva pūrvoktā nītiravatarati //
For here too, in the expression "na na-evam" [not not-thus], there is negation of negation, while "evam" remains in its own unnegated form; thus the same reasoning mentioned before applies here as well.
etat sarvaṃ kumāriloktam upanyastam, sāmprataṃ sarvaśabdasyetyādinodyotakāroktam apohadūṣaṇam āśaṅkate
All this has been presented as stated by Kumārila; now [the author] considers Uddyotakara's refutation of apoha [beginning with] "in the case of the word 'sarva'" etc.
sarvaśabdasya kaścārtho vyavacchedyaḥ prakalpyate / nāsarvanāma kiñciddhi bhavedyasya nirākriyā //
What meaning is assumed to be excluded in the case of the word "sarva" [all]? For there is nothing that could be [called] "non-all" which could be excluded.
ekādyasarvamiti cedarthāpohaḥ prasajyate / aṅgānāṃ pratiṣiddhatvād aniṣṭeścāṅginaḥ pṛthak //
If it is argued that "one etc." is [what constitutes] "non-all," then this would result in the exclusion of the meaning itself, because the parts would be excluded and no whole is accepted separately.
evaṃ samūhaśabdārthe samudāyivyapohataḥ /
Similarly, regarding the meaning of the word "samūha" [group], [there would be] exclusion of the constituents.
anyāniṣṭeś ca sarve 'pi prāpnuvanti nirarthakāḥ //
And since nothing else is accepted [as meaning], all [such words] would become meaningless.
dvyādiśabdā iheṣṭāś ca ye samuccayagocarāḥ / ekādipratiṣedhena na bhaveyus tathāvidhāḥ //
The words [denoting] 'two' and the rest, which are accepted here as referring to collections, could not be of such a nature due to the exclusion of [numbers like] 'one' and others.
nāgaur iti ca yo 'pohā gośabdasyārtha ucyate /
And when the meaning of the word 'go' [cow] is said to be the exclusion of 'non-cow',
sa kiṃ bhāvo 'thavābhāvo bhāvo gaurvāthavāpyagauḥ //
is that [meaning] positive or negative? If positive, is it the cow or the non-cow?
gauścennāsti vivādo 'yam arthastu vidhilakṣaṇaḥ / agaurgośabdavācyaścedatiśabdārthakauśalam //
If it is the cow, there is no dispute, as the meaning is of positive nature. [But] if the word 'go' denotes non-cow, [that would show] extraordinary skill in [understanding] word meanings!
abhāvo 'pi na yukto 'yaṃ praiṣādīnām asambhavāt /
This negative [meaning] too is not logical, because injunctions and such would be impossible.
na hi gośabdataḥ kaścid abhāvaṃ pratipadyate //
For nobody comprehends a negation from the word 'go' [cow].
sa hyāha apohaḥ śabdārtha ityayuktam etat avyāpakatvāt /
He [Uddyotakara] indeed says: "It is incorrect [to say] that apoha is the meaning of words, because it is not all-pervading."
yatra dvairāśyaṃ bhavati tadetarapratiṣedhāditaratpratīyate yathā gaur itipade gauḥ pratīyamāno 'gaur niṣedhyamānaḥ /
Where there is a dichotomy, one thing is understood through the negation of its opposite, as in the case of the word 'go', where 'cow' is understood [and] 'non-cow' is negated.
na punaḥ sarvapade etad asti / na hyasarvaṃ nāma kiñcid asti yat sarvaśabdena nivartyeta /
But this is not [the case] with the word 'sarva' [all], for there exists nothing called 'non-all' which could be negated by the word 'sarva'.
atha manyase ekādyasarvaṃ tatsarvaśabdena nivartyata iti /
If you think that 'one' etc., [being] non-all, are negated by the word 'sarva', [then...]
tan na svārthāpavādadoṣaprasaṅgāt
This [view] is not [correct], as it would involve the fault of words abandoning their own meaning.
evaṃ hyekādivyudāsena pravartamānaḥ sarvaśabdo 'ṅgapratiṣedhādaṅgavyatiriktasyāṅgino 'nabhyupagamād anarthakaḥ syāt
For if the word 'sarva' [all] were to function by excluding 'one' etc., because of the negation of [its] parts and the non-acceptance of a whole distinct from [its] parts, [the word] would become meaningless.
aṅgaśabdenaikadeśa ucyate
By the word 'aṅga' [here], a part is meant.
evaṃ sarve samudāyaśabdā ekadeśapratiṣedharūpeṇa pravartamānāḥ samudāyivyatiriktasyānyasya samudāyasyānabhyupagamād anarthakāḥ prāpnuvanti
Similarly, all collective words, if functioning in the form of negating [their] parts, would become meaningless, since no collection distinct from [its] collected members is accepted.
dvyādiśabdānāṃ ca viṣayasamuccayatvād ekādipratiṣedhe pratiṣidhyamānānamārthānām asamuccayatvād anarthakatvam taddarśayati na bhaveyus tathāvidhā iti
And since words like 'two' etc. refer to collections, if they were to negate 'one' etc., they would become meaningless due to the impossibility of collecting the negated meanings, [which] is shown [by the statement] "they would not be of that nature."