A word of explanation is necessary as regards my interpretation of the Sa@mkhya-Yoga system. The _Sa@mkhya karika_ is the oldest Sa@mkhya text on which we have commentaries by later writers. The _Sa@mkhya sutra_ was not referred to by any writer until it was commented upon by Aniruddha (fifteenth century A.D.). Even Gu@naratna of the fourteenth century A D. who made allusions to a number of Sa@mkhya works, did not make any reference to the _Sa@mkhya sutra_, and no other writer who is known to have flourished before Gu@naratna seems to have made any reference to the _Sa@mkhya sutra_. The natural conclusion therefore is that these sutras were probably written some time after the fourteenth century. But there is no positive evidence to prove that it was so late a work as the fifteenth century. It is said at the end of the _Sa@mkhya karika_ of Is'varak@r@s@na that the karikas give an exposition of the Sa@mkhya doctrine excluding the refutations of the doctrines of other people and excluding the parables attached to the original Sa@mkhya works--the _@Sa@s@titantras'astra_. The _Sa@mkhya sutras_ contain refutations of other doctrines and also a number of parables. It is not improbable that these were collected from some earlier Sa@mkhya work which is now lost to us. It may be that it was done from some later edition of the _@Sa@s@titantras'astra_ (_@Sa@s@titantroddhara_ as mentioned byGu@naratna), but this is a mere conjecture. There is no reason to suppose that the Sa@mkhya doctrine found in the sutras differs in any important way from the Sa@mkhya doctrine as found in the _Sa@mkhya karika_. The only point of importance is this, that the _Sa@mkhya sutras_ hold that when the Upani@sads spoke of one absolute pure intelligence they meant to speak of unity as involved in the class of intelligent puru@sas as distinct from the class of the gu@nas. As all puru@sas were of the nature of pure intelligence, they were spoken of in the Upani@sads as one, for they all form the category or class of pure intelligence, and hence may in some sense be regarded as one. This compromise cannot be found in the _Sa@mkhya karika_. This is, however, a case of omission and not of difference. Vijnana Bhik@su, the commentator of the _Sa@mkhya sutra_, was more inclined to theistic Sa@mkhya or Yoga than to atheistic Sa@mkhya. This is proved by his own remarks in his _Samkhyapravacanabha@sya, Yogavarttika_, and _Vijnanam@rtabhasya_ (an independent commentary on the Brahmasutras of Badarayana on theistic Sa@mkhya lines). Vijnana Bhiksu's own view could not properly be called a thorough Yoga view, for he agreed more with the views of the Sa@mkhya doctrine of the Pura@nas, where both the diverse puru@sas and the prak@rti are said to be merged in the end in Is'vara, by whose will the creative process again began in the prakrti at the end of each pralaya. He could not avoid the distinctively atheistic arguments of the _Sa@mkhya sutras_, but he remarked that these were used only with a view to showing that the Sa@mkhya system gave such a rational explanation that even without the intervention of an Is'vara it could explain all facts. Vijnana Bhik@su in his interpretation of Sa@mkhya differed on many points from those of Vacaspati, and it is difficult to say who is right. Vijnana Bhik@su has this advantage that he has boldly tried to give interpretations on some difficult points on which Vacaspati remained silent. I refer principally to the nature of the conception of the gu@nas, which I believe is the most important thing in Sa@mkhya. Vijnana Bhik@su described the gu@nas as reals or super-subtle substances, but Vacaspati and Gau@dapada (the other commentator of the _Sa@mkhya karika_) remained silent on the point. There is nothing, however, in their interpretations which would militate against the interpretation of Vijnana Bhik@su, but yet while they were silent as to any definite explanations regarding the nature of the gu@nas, Bhik@su definitely came forward with a very satisfactory and rational interpretation of their nature.
Since no definite explanation of the gu@nas is found in any other work before Bhik@su, it is quite probable that this matter may not have been definitely worked out before. Neither Caraka nor the _Mahabharata_ explains the nature of the gu@nas. But Bhik@su's interpretation suits exceedingly well all that is known of the manifestations and the workings of the gu@nas in all early documents. I have therefore accepted the interpretation of Bhik@su in giving my account of the nature of the gu@nas. The _Karika_ speaks of the gu@nas as being of the nature of pleasure, pain, and dullness (_sattva, rajas_ and _tamas_). It also describes sattva as being light and illuminating, rajas as of the nature of energy and causing motion, and tamas as heavy and obstructing. Vacaspati merely paraphrases this statement of the _Karika_ but does not enter into any further explanations. Bhik@su's interpretation fits in well with all that is known of the gu@nas, though it is quite possible that this view might not have been known before, and when the original Sa@mkhya doctrine was formulated there was a real vagueness as to the conception of the gu@nas.
There are some other points in which Bhik@su's interpretation differs from that of Vacaspati. The most important of these may be mentioned here. The first is the nature of the connection of the buddhi states with the puru@sa. Vacaspati holds that there is no contact (_sa@myoga_) of any buddhi state with the puru@sa but that a reflection of the puru@sa is caught in the state of buddhi by virtue of which the buddhi state becomes intelligized and transformed into consciousness. But this view is open to the objection that it does not explain how the puru@sa can be said to be the experiencer of the conscious states of the buddhi, for its reflection in the buddhi is merely an image, and there cannot be an experience (_bhoga_) on the basis of that image alone without any actual connection of the puru@sa with the buddhi. The answer of Vacaspati Mis'ra is that there is no contact of the two in space and time, but that their proximity (_sannidhi_) means only a specific kind of fitness (_yogyata_) by virtue of which the puru@sa, though it remains aloof, is yet felt to be united and identified in the buddhi, and as a result of that the states of the buddhi appear as ascribed to a person. Vijnana Bhik@su differs from Vacaspati and says that if such a special kind of fitness be admitted, then there is no reason why puru@sa should be deprived of such a fitness at the time of emancipation, and thus there would be no emancipation at all, for the fitness being in the puru@sa, he could not be divested of it, and he would continue to enjoy the experiences represented in the buddhi for ever. Vijnana Bhik@su thus holds that there is a real contact of the puru@sa with the buddhi state in any cognitive state. Such a contact of the puru@sa and the buddhi does not necessarily mean that the former will be liable to change on account of it, for contact and change are not synonymous. Change means the rise of new qualities. It is the buddhi which suffers changes, and when these changes are reflected in the puru@sa, there is the notion of a person or experiencer in the puru@sa, and when the puru@sa is reflected back in the buddhi the buddhi state appears as a conscious state. The second, is the difference between Vacaspati and Bhik@su as regards the nature of the perceptual process. Bhik@su thinks that the senses can directly perceive the determinate qualities of things without any intervention of manas, whereas Vacaspati ascribes to manas the power of arranging the sense-data in a definite order and of making the indeterminate sense-data determinate. With him the first stage of cognition is the stage when indeterminate sense materials are first presented, at the next stage there is assimilation, differentiation, and association by which the indeterminate materials are ordered and classified by the activity of manas called sa@mkalpa which coordinates the indeterminate sense materials into determinate perceptual and conceptual forms as class notions with particular characteristics. Bhik@su who supposes that the determinate character of things is directly perceived by the senses has necessarily to assign a subordinate position to manas as being only the faculty of desire, doubt, and imagination.
It may not be out of place to mention here that there are one or two passages in Vacaspati's commentary on the _Sa@mkhya karika_ which seem to suggest that he considered the ego (_aha@mkara_) as producing the subjective series of the senses and the objective series of the external world by a sort of desire or will, but he did not work out this doctrine, and it is therefore not necessary to enlarge upon it. There is also a difference of view with regard to the evolution of the tanmatras from the mahat; for contrary to the view of _Vyasabha@sya_ and Vijnana Bhik@su etc. Vacaspati holds that from the mahat there was aha@mkara and from aha@mkara the tanmatras [Footnote ref 340]. Vijnana Bhik@su however holds that both the separation of aha@mkara and the evolution of the tanmatras take place in the mahat, and as this appeared to me to be more reasonable, I have followed this interpretation. There are some other minor points of difference about the Yoga doctrines between Vacaspati and Bhik@su which are not of much philosophical importance.