Yoga Sutra - Samādhi Pada 1.11
अनुभूतविषयासंप्रमोषः स्मृतिः ॥११॥
anubhūta-viṣaya-asaṃpramoṣaḥ smṛtiḥ ||11||
Memory is the recollection of an experienced condition.
Memory is the retention of [images of] sense objects that have been experienced.
Vyāsa notes that memories are generated from and thus dependent on the other types of vṛttis described in the preceding sūtras: right knowledge, error, imagination, and sleep (one has memories of objects of knowledge, error, etc.). Therefore, says Vācaspati Miśra, memory is mentioned last on the list of vṛttis. Nonetheless, the Yoga school considers memory a vṛtti in its own right. Patañjali here describes memory as the retention or, more literally, the not slipping away, asampramoṣa, of an object of experience, anubhūta viṣaya. As noted, every object that has ever been experienced forms a saṁskāra, an imprint, in the citta mind. The mind forms an impression of an object, called a pratyaya, through the sense organs. As noted in the previous sūtra, a pratyaya is the specific content of a vṛtti, a unitary image like the individual still of a movie, where a vṛtti is more a sequence and thus a series of pratyayas, comparable to the reel itself containing the series of stills. Once this pratyaya or active image of this object is no longer of interest to the mind, it becomes an inactive, or latent, saṁskāra. Thus vṛttis, and their pratyaya content, are retained as saṁskāras when they fade.
A person approaching a red rose, for example, receives an impression through the senses of the sight and smell of a red object of a particular shape with a certain odor. This input is recognized by and imprinted on the mind as a rose, a fragrant member of the flower family. However, even after the mind has moved on to other things, this cognition or impression remains embedded in the citta mind in the form of a saṁskāra, imprint. Since the mind is exposed to numerous sense objects all the time, and has been for numerous lifetimes, there are unlimited saṁskāras continually being embedded in the citta, which are all potentially retrievable. Memory, then, the fifth and final type of vṛtti listed by Patañjali, consists of the retrieval of these saṁskāras; memories are the reactivation of the imprints of sense objects that one has experienced and recognized in the past. (Vijñānabhikṣu makes a distinction here between memory and recognition.) When these saṁskāras do not slip away (fall into the unconscious), or, in Sāṅkhyan categories, when they do not become overly covered by tamas, they constitute memory. The citta can perhaps be compared to a lake, and saṁskāra memories to the pebbles at the bottom of the lake. If the lake is peaceful and crystal clear (sāttvic), the pebbles are easily visible; if it is choppy (rājasic), less so; and if murky (tāmasic), the pebbles become very hard or impossible to perceive.
Vyāsa states that memory is of two types—real, namely, recollection of things that actually happened; and imagined, such as in dreams, which involve the spontaneous and more random activation of saṁskāras but are still memories of sorts. Bhoja Rāja says that the saṁskāras that activate during dreams do so because of the force of the impression that created them. In other words, they were vivid events. On this topic, Rāmānanda Sarasvatī raises the question that if all saṁskāras consist of, and only of, events or objects that have been actually experienced, then how can we account for those dreams where one might imagine oneself, for example, as having the body of an elephant? What saṁskāras produced these? Surely one has not had such an experience in real life. Such memories, he answers, are the result of error, which affects the dream state just as it does the waking state.
One might add that one cannot dream of, or for that matter even imagine, something that does not exist as a saṁskāra or set of saṁskāras in one’s citta, and saṁskāras, in turn, correspond to something one has actually experienced. However, saṁskāras can be clustered together in a way that corresponds to something that has never been experienced in reality, such as a flying elephant or the horns of a hare. What is taking place in such imaginative instances is the merging together of two sets of actual saṁskāras; in other words, one set of memory experiences, recollections of the act of flying or of horns, are superimposed on and blended with other memory imprints, those of an elephant or a hare. (As a point of interest, in addition to the fantasy nature of most dreams, Vedānta Sūtra III.2.4 allows that some dreams can also serve as omens of real events.)
The Jains accept memory, smṛti, as a pramāṇa (I.6) in its own right, but this is rejected by the Hindu schools since memory does not present immediate or new knowledge, which pramāṇa for these philosophical schools must do, but re-presents something that has already happened in the past. However, smṛti is used by Hindus to refer to a category of āgama, sacred text, which is a pramāṇa. This category of smṛti includes in practice more or less everything other than the śruti, or the Vedic corpus—the epics, Purāṇas, and sūtra traditions including Patañjali’s Yoga Sūtras. However, since smṛti sacred texts present themselves, at least nominally, as dependent on or derivative from the śruti for their authoritativeness, smṛti texts are not considered pramāṇa on the same level as śruti in orthodox discourse, at least not officially (although in practice, it has been the smṛti that has determined the beliefs and practices of Hindus throughout the ages). Or, put differently, as in the Vedānta tradition, smṛti can be accepted as āgama-pramāṇa to the extent it does not conflict with the śruti.
This concludes the discussion on what constitutes a vṛtti. According to Vyāsa, the five types of vṛttis identified by Patañjali are either experienced as pleasurable in nature, resulting in attachment (sattva); unpleasurable, resulting in aversion (rajas); or deluded, resulting in ignorance (tamas). Of course, as Vijñānabhikṣu notes, from the absolute perspective pleasure also results in pain, because pleasure generates attachment to the object of pleasure, and this attachment ultimately results in pain upon the loss of this object; attachment also perpetuates saṁsāra, as we will discover in II.2ff.
One might add here that Patañjali indicated in I.5 that these five vṛttis are either kliṣṭa, detrimental, or akliṣṭa, beneficial, conducive to the goal of yoga. Thus, for example, the vṛtti of sleep might be considered nondetrimental when it is not excessive but simply adequate for the well-being of the body, but detrimental when in excess of this; the vṛtti of memory may be nondetrimental when one keeps the goals of yoga in mind, but detrimental when one harps back on past sensual indulgences. Perhaps the hardest vṛtti to envision in this regard is how error could be sāttvic and akliṣṭa, conducive to the goal of yoga, but irrespective of the state of one’s citta, if the instruments of the senses are defective error may nonetheless occur. For example, if he has poor eyesight, even the accomplished yogī will still perceive things erroneously and thus be subject to error. One can play around creatively with other possible scenarios of error being akliṣṭa: One might walk into a yoga studio taught by dedicated teachers grounded in Patañjali’s teachings, mistaking it for a health spa, but, once there, become inspired to undertake the practices and philosophy of the system, and thus one’s initial error becomes conducive to the goals of the yoga. There are certainly plenty of examples in the Purāṇas that might illustrate error from a bhakti, devotional, perspective.
“In the Bhāgavata tradition, the residents of Vraj, where Kṛṣṇa spent his childhood, unaware of Kṛṣṇa’s divinity, mistook him to be their son, or friend, or lover, etc., but such error is highly favorable to the goal in this Yoga tradition and fundamental to its theology of līlā, the pastimes of the incarnation of God in the world (as well as indicative of the extraordinary past-life yogic attainments of those residents, which allowed them this opportunity of such intimate association with God in this form). Whatever their nature, says Vyāsa, vṛttis must be restrained for any type of concentrative state—either samprajñāta or asamprajñāta—to take place, as Patañjali announced at the beginning of his sūtras.