Atman, from Sanskrit ātman, is a term for the self that occupies a central place in Hinduism and in the wider history of Indian philosophy. Depending on its context, it can mean oneself, the living person, the enduring subject of experience, or the deepest reality of a being. Its philosophical significance lies in questions about what remains identical through bodily and mental change, what experiences the consequences of action, and what, if anything, can become free from suffering and repeated birth. “Soul” is a common English translation, but it can introduce assumptions about individuality, creation, and the relation between mind and body that particular Indian theories do not share.
There is no single doctrine of atman accepted by all Hindu traditions. The Upanishads connect inquiry into the self with inquiry into brahman, ultimate reality, but their formulations admit different interpretations. Some schools identify the true self with brahman; others maintain an enduring distinction between individual selves and God. Other Indian systems understand selves without making their identity with brahman a central claim. Buddhist critiques of a permanent self, and Jain accounts of individual living souls, developed within this broader field of debate.
Atman is therefore simultaneously a linguistic expression, a subject of metaphysics, and a focus of religious discipline. Its history involves changing interpretations of scripture, arguments about memory and consciousness, theories of moral continuity, and practices intended to distinguish the self from what is mistakenly taken to be oneself.
Meaning and early Vedic background
The Sanskrit noun ātman has a wider semantic range than the technical expression “eternal soul.” In ordinary and philosophical usage it can function reflexively, meaning “oneself,” or designate a person, the body, an essential nature, or a principle of life. The nominative singular form is ātmā; ātman is the stem conventionally used in English scholarship. Its historical etymology is debated, although its early associations include breath and vital existence. A translation must therefore attend to context rather than assume that every occurrence expresses a developed metaphysical theory.
In the Rigveda, the oldest collection of Vedic hymns, the word appears without the systematic account of selfhood found in later literature. A notable example is Rigveda 1.115.1, which calls the sun the ātman of what moves and what stands still. Here the term suggests an animating or essential principle on a cosmic scale. Such passages show that the vocabulary of inward life and the vocabulary of cosmic order were already capable of overlapping; they do not establish that the Rigveda taught the later Advaita doctrine of the identity of self and brahman.
The development of this vocabulary occurred within the religious world of the Vedas, where sacrifice, sacred speech, bodily processes, and the structure of the universe were related through networks of correspondence. The Brāhmaṇa and Āraṇyaka texts connect aspects of the sacrificial performance with the sacrificer and with cosmic powers. Breath, speech, sight, hearing, and mind could each be considered indispensable to life, and narratives sometimes stage contests among these faculties. Such discussions helped create the conceptual setting in which a question could arise about the principle that supports them all.
The transition from these materials to the Upanishads was not a simple replacement of “external ritual” by “internal spirituality.” Ritual speculation and inquiry into the self frequently coexist in the same textual traditions. A sacred act could be interpreted as having an inward counterpart, while knowledge of a correspondence could be said to alter the significance or efficacy of its performance. The self became a point at which the body, sacrificial order, and cosmos could be investigated together.
This background matters because later commentators often read early passages through much more systematic doctrines. A Vedic reference to a person’s ātman need not designate an immutable witness, and a description of a cosmic self need not deny the reality of individual persons. Historically, the term supplied a flexible vocabulary from which competing theories could develop, rather than naming one fully formed doctrine from its earliest appearances.
The self in the principal Upanishads
The early principal Upanishads, particularly the Bṛhadāraṇyaka and Chāndogya, contain some of the most influential discussions of atman. These works were composed and transmitted over extended periods during the first millennium BCE; their exact chronology is uncertain. They present dialogues, teaching narratives, analogies, and speculative identifications rather than a single continuous philosophical system. Questions about the self arise alongside questions about death, desire, knowledge, and the ground of the world.
In the Bṛhadāraṇyaka Upanishad, Yājñavalkya’s instruction to Maitreyī makes knowledge of the self central to understanding what is valued. At 2.4.5, repeated in a parallel passage at 4.5.6, the self is said to be what should be seen, heard about, reflected upon, and deeply contemplated. The teaching connects the value of apparently separate objects and relationships with the self for whose sake they are dear. It is not simply a recommendation to prefer personal advantage: the inquiry progressively unsettles an ordinary distinction between the individual knower and independently possessed objects.
Other passages emphasize the limits of treating the self as an object. Bṛhadāraṇyaka 3.4.2 asks, in effect, how one could see the seer of seeing or know the knower of knowing. The formulation later became important for accounts of the self as the presupposition of experience rather than one more experienced thing. The expression neti neti, “not this, not this,” at 2.3.6 likewise resists exhaustive identification of ultimate reality with particular descriptions. Different commentators disagree about whether such language excludes all attributes or only inadequate, limiting ones.
The Chāndogya Upanishad develops another influential approach through Uddālaka Āruṇi’s instruction to Śvetaketu. Examples involving clay, a seed, and salt dissolved in water lead toward the claim that the subtle reality underlying things is also the self. The formula tat tvam asi, conventionally translated “you are that,” recurs in chapter 6. Its interpretation became a major point of disagreement among later Vedānta schools: nondualists read it as revealing identity, while other interpreters explain it through dependence, qualification, or a shared divine ground.
Later Upanishads offer additional models. The Kaṭha compares the body to a chariot, the intellect to its driver, the mind to the reins, and the self to the chariot’s lord. The Taittirīya describes progressively interior dimensions associated with food, breath, mind, understanding, and bliss, subsequently interpreted through the teaching of five sheaths. The Māṇḍūkya examines waking, dreaming, and deep sleep before presenting a “fourth,” turīya, not reducible to these ordinary conditions. These accounts share an effort to distinguish the deepest self from immediately apparent identity, but they do so through different conceptual and pedagogical structures.
Action, rebirth, and the Bhagavad Gita
Atman became closely connected with accounts of karma, saṃsāra, and moksha. Karma concerns action and its consequences; saṃsāra is the recurring condition of birth and death; moksha is liberation from that condition. A theory of an enduring self can explain how the consequences of action belong to a being whose present body differs from a previous embodiment. Nevertheless, the precise relation between self, action, and continuity varies considerably across schools.
The Bhagavad Gita, a dialogue between Krishna and Arjuna within the Mahābhārata, gives one of the most influential presentations of the imperishable self. In chapter 2, Krishna contrasts the changing body with that which is neither born nor destroyed. Verse 2.20 denies birth and death of the self, while 2.22 compares the replacement of bodies to the replacement of worn garments. The accompanying claim that weapons cannot cut it and fire cannot burn it expresses its difference from destructible material things.
These teachings occur within a narrative about action, not withdrawal from all activity. Arjuna’s crisis concerns participation in a war involving relatives and teachers. Krishna’s response combines an account of the imperishable self with arguments about duty, disciplined action, knowledge, and devotion. The immortality teaching cannot be detached from this setting and treated as the text’s sole account of moral responsibility. The Gita does not generally infer that bodily suffering or the consequences of conduct are insignificant because the self is imperishable.
The term ātman itself retains several meanings within the work. In chapter 6, the instruction to raise oneself by oneself, followed by the statement that the self can be its own friend or enemy, concerns disciplined relations among dimensions of a person rather than an opposition between two independently existing souls. Context determines whether “self” refers primarily to the person, the mind under discipline, or the deeper spiritual principle.
The Gita also distinguishes the embodied individual from the supreme divine reality without fixing every such distinction in a way accepted by later interpreters. Chapter 13 introduces the “field,” kṣetra, and the “knower of the field,” kṣetrajña, while identifying Krishna as the knower in all fields. Chapter 15 describes the living being as an eternal portion of Krishna. Nondualist, qualified-nondualist, and dualist commentators offer different explanations of these formulations.
The resulting tradition links self-knowledge with several modes of religious life. Action performed without possessive attachment, contemplation, and devotion may all contribute to liberation, but their relative roles differ by interpretation. The Gita thus provided later thinkers with both a vocabulary of the enduring self and a framework for relating that self to practical obligation.
Atman and brahman in Advaita Vedanta
Advaita Vedānta holds that the true self is non-different from brahman. Its most influential early systematic exponent was Śaṅkara, active in approximately the eighth century CE, although nondual interpretations preceded him. His commentaries on the principal Upanishads, the Gita, and the Brahma Sūtras established a durable framework for interpreting apparently diverse scriptural teachings as culminating in nonduality.
The central distinction is between the self as pure consciousness and the empirical individual who identifies with a particular body and mind. The latter is often called the jīva, the embodied living individual. Ordinary experience attributes bodily features, mental conditions, agency, and vulnerability to “I.” Advaita explains this through adhyāsa, superimposition: properties belonging to the body or mind are attributed to the self, while the self’s consciousness is attributed to what is otherwise an object of awareness. Śaṅkara’s introduction to his Brahma Sūtra commentary makes this mutual attribution foundational to ordinary experience.
On this account, atman is not a hidden object located somewhere inside the body. It is that by which bodily sensations, thoughts, perceptions, and their absence are known. It is described as self-revealing because its presence does not require a further act of cognition to illuminate it. The witness, sākṣin, is a common explanatory designation, but Advaita ultimately denies that the self stands in an irreducible relation to a second, independently real entity.
The distinction between empirical and ultimate standpoints is essential. At the empirical level, persons act, acquire knowledge, incur karmic consequences, and pursue liberation. At the ultimate level, brahman is nondual, and the self has never actually become bound. Ignorance, avidyā, explains the appearance of limitation; knowledge removes that ignorance rather than producing a previously nonexistent freedom. Calling the world mithyā means that it lacks independent, ultimate reality, not simply that ordinary experience is equivalent to nothing.
Advaita’s interpretation of Upanishadic identity statements therefore differs from the proposition that an individual personality is an omnipotent deity. Features that make an individual this particular person—biography, body, memory, social position—are not what the teaching identifies with brahman. The claimed identity concerns consciousness free from limiting identifications.
Later Advaitins debated the location and structure of ignorance, the status of the individual, and the relation between knowledge and contemplative practice. These disagreements show that even within nondualism, the formula “atman is brahman” did not eliminate philosophical problems. It generated detailed inquiry into how error is possible and how a teaching expressed through language can disclose a reality said not to be an ordinary object.
Individual selves in theistic Vedanta
Other schools of Vedānta accept the authority of the same foundational texts while rejecting Advaita’s interpretation of their final meaning. They generally affirm enduring individual selves and understand liberation through a relation to a supreme personal reality. Their disagreements concern whether difference is ultimate, how finite consciousness depends on God, and whether liberation preserves agency and individuality.
Rāmānuja, who lived in the eleventh and twelfth centuries, developed the influential system known as Viśiṣṭādvaita, or qualified nondualism. Brahman is one, but that unity includes real differentiation. Conscious selves and nonconscious reality stand to God in a relation often explained through the analogy of body and indwelling self. They depend upon divine support and direction without becoming unreal appearances. Thus the unity of reality does not require the elimination of individual subjects.
In this system, the individual self is an enduring conscious subject, not identical with its body or with a succession of thoughts. Rāmānuja distinguishes the self from its attributive consciousness, through which objects are known. In embodied existence, the manifestation of this knowledge is restricted by karma; in liberation it is no longer subject to those restrictions. The liberated self retains its identity and participates in the experience and service of God rather than discovering that individuality was wholly illusory.
Madhva, active in the thirteenth century, advanced Dvaita Vedānta, a stronger account of real difference. God, identified with Vishnu, is independent; individual selves and material reality are dependent. Difference between God and a self, between one self and another, and between consciousness and matter is not merely an effect of ignorance. Madhva’s tradition also teaches differences among selves in their capacities and destinies. Liberation does not erase the distinction between the worshipper and the supreme being.
Other Vedānta traditions offer further arrangements of unity and difference. Nimbārka’s difference-and-nondifference account, Vallabha’s pure nondualism, and the Gauḍīya Vaiṣṇava teaching of inconceivable difference and nondifference each articulate a distinctive relation between individual consciousness and the divine. These positions cannot be adequately classified as either a Western-style individual soul doctrine or a simple identity of everything with an impersonal absolute.
For these traditions, Vaiṣṇava devotion is not merely a preliminary technique destined to disappear when higher knowledge is attained. Loving relation, dependence, grace, and service can belong to the final condition itself. The question “What is atman?” is consequently inseparable from another: what kind of relationship between the finite self and the supreme reality can persist in liberation?
Selfhood in other Hindu philosophical systems
Theories of atman extend beyond Vedānta. Nyāya and Vaiśeṣika defend enduring selves within a broadly realist account of the world. Selves are substances distinct from bodies, senses, and the internal mental organ, manas. Cognition, desire, aversion, pleasure, pain, and effort are treated as qualities belonging to a self rather than as independent subjects. Classical accounts generally recognize many selves, each associated with its own history of action and experience.
Nyāya philosophers argue for a self through phenomena such as recognition, recollection, and the coordination of information from different senses. Recognizing an object as one previously encountered appears to connect present cognition with an earlier experience. A single continuing subject is proposed to account for that connection. Likewise, seeing an object and later identifying it by touch is used to argue that the subject is not identical with any one sensory faculty. These are inferences within a systematic epistemology, not simply appeals to scriptural authority.
Importantly, the self in classical Nyāya is not always conscious in the manner asserted by Advaita. Cognitions arise under appropriate conditions involving self, mind, senses, and objects. Consciousness is therefore not necessarily an uninterrupted intrinsic manifestation. Classical descriptions of liberation as the cessation of suffering and of ordinary mental qualities differ sharply from accounts of liberation as the unrestricted manifestation of blissful consciousness, although later authors developed varied positions.
Sāṃkhya uses puruṣa as its principal term for conscious selfhood. It distinguishes a plurality of puruṣas from prakṛti, the productive material principle from which the intellect, ego, mind, senses, and physical elements develop. The psychological apparatus belongs to nature, not to pure consciousness. Bondage occurs because the activity of that apparatus is mistakenly attributed to the conscious subject; liberation comes through discriminating their difference. Unlike Advaita, classical Sāṃkhya does not reduce all conscious subjects to one brahman.
Classical Yoga, represented by Patañjali’s Yoga Sūtras, adopts much of this distinction. Yoga Sūtra 1.3 states that when the fluctuations of the mind are stilled, the seer abides in its own nature. Its final goal, kaivalya, is the isolation or independence of consciousness from the processes of nature. Mīmāṃsā, meanwhile, defends an enduring self in connection with Vedic obligation, agency, and the results of action, while its major schools disagree about the structure of cognition and self-awareness. These systems demonstrate that acceptance of a self does not by itself imply nonduality, theism, or a single account of consciousness.
Buddhist critiques and Jain accounts
Buddhism made the rejection of an independently enduring self a central point of disagreement with many other Indian traditions. The doctrine of anattā, Sanskrit anātman, is commonly rendered “not-self” or “no-self.” Early Buddhist analyses examine bodily form, feeling, perception, volitional formations, and consciousness—the five aggregates—and deny that any is properly regarded as an unchanging, sovereign self. Their instability and susceptibility to suffering undermine claims that they are “mine,” “I,” or “my self” in the liberating sense sought by the teachings.
This position does not amount to denying that people speak, act, remember, or bear consequences. Buddhist accounts explain continuity through causally connected processes rather than through an immutable bearer underlying them. Dependent origination supplies a framework in which events arise through conditions. Later Buddhist philosophers developed different accounts of momentariness, cognition, and conventional personhood, but generally resisted a permanent self independent of the aggregates.
Debates with self-theorists often focused on memory and agency. A defender of atman might contend that the same subject must both perform an action and later experience its result. A Buddhist response can distinguish numerical identity of a permanent substance from causal continuity within a stream of events. Conversely, Hindu critics questioned whether a sequence of distinct events could adequately explain ownership of memory or responsibility. Neither side simply denied the practical phenomena; they disagreed about the explanatory structure those phenomena required.
Some Mahāyāna scriptures employ affirmative language associated with buddha-nature, including, in certain texts, the vocabulary of self. This has produced substantial interpretive debate. Such language cannot automatically be identified with the atman of Vedānta: many Buddhist explanations understand it in relation to emptiness, freedom from defilement, or the potential for awakening, rather than as an independently existing personal substance. Its meaning must be assessed within particular texts and commentarial traditions.
Jainism, by contrast, explicitly affirms a plurality of enduring jīvas. A jīva is characterized by consciousness and is distinguished from nonliving reality. Jain accounts explain bondage through karmic matter associated with the soul as a result of activity and passion. Liberation removes this bondage and permits the manifestation of the soul’s knowledge and other capacities. Liberated souls remain distinct; they do not merge into a single universal self.
The Indian debate is therefore not adequately represented as Hindu belief in a soul opposed to a uniform alternative. Buddhist traditions offer several process-based accounts of continuity, Jainism affirms real individual souls through a distinctive theory of karmic materiality, and Hindu schools disagree among themselves about what a self is and how its persistence should be explained.
Knowing the self and attaining liberation
The claim that the self is different from body and mind raises an epistemological problem: how can it be known? Indian philosophers classify reliable ways of knowing as pramāṇas, but disagree about their number, scope, and application. Perception, inference, and authoritative testimony are especially important in discussions of atman. A school may accept the self as immediately evident, infer it from experience, or hold that scripture is indispensable for discovering its ultimate nature.
Advaita distinguishes the undeniable presence of awareness from correct knowledge of what that awareness is. Ordinary persons already use “I,” yet identify the self with characteristics that belong to body or mind. Upanishadic instruction is held to remove this misidentification. It does not create the self or reveal it as a new sensory object. The familiar sequence of hearing the teaching, reasoning about it, and sustained contemplation—śravaṇa, manana, and nididhyāsana—organizes the process of overcoming misunderstanding and contrary habits.
The role of meditation consequently differs across systems. In Advaita, a temporarily quiet mind is not necessarily liberating knowledge: mental stillness can occur without removal of ignorance. In classical Yoga, disciplined concentration belongs to a detailed transformation of mental activity culminating in discriminative knowledge and independence of the seer. In devotional Vedānta, recollection, worship, and contemplation are related to knowledge of the self’s dependence on God. Similar outward exercises can therefore serve different philosophical ends.
Deep sleep provides a particularly revealing disputed case. Upanishadic and Advaitic discussions use the absence of ordinary objects and individual differentiation in sleep to investigate consciousness beyond waking identity. Later Advaitins appeal to reports of having slept happily and having known nothing. Their opponents dispute what such reports establish: whether awareness persisted, whether a memory can concern an absence, and whether inference rather than direct recollection explains the later judgment. The shared phenomenon does not compel one metaphysical interpretation.
Liberation also depends on how bondage is defined. If bondage is ignorance of an already nondual self, knowledge removes an error. If it is the association of a real soul with karmic effects, liberation requires the termination of that association. If it involves dependence distorted by egoism and attachment, devotion and divine grace may be central. Accordingly, atman theories are connected with ethics and discipline without yielding a single universal program.
Teachings about seeing the self in all beings can support the rejection of hatred and the cultivation of nonviolence, as in the opening verses of the Īśā Upanishad. Nevertheless, a metaphysical claim about selfhood and a specific moral prescription remain distinguishable. Historical traditions articulate their connection through additional teachings about duty, desire, conduct, and the qualifications required for liberating knowledge.
Modern interpretation and comparative philosophy
The modern international reception of atman was shaped by translation, colonial scholarship, Hindu reform, and new forms of global religious exchange. European translations often used “soul,” “spirit,” or “self,” each carrying an intellectual history of its own. Anquetil-Duperron’s Latin Oupnek’hat, published in 1801–1802 from a Persian translation of Upanishadic materials, was an important route through which these teachings reached European readers. Arthur Schopenhauer encountered the Upanishads through this mediated textual tradition and drew comparisons with his own distinction between appearance and underlying reality.
In the nineteenth and twentieth centuries, figures such as Swami Vivekananda presented Vedāntic accounts of the self to audiences outside India. His participation in the 1893 World’s Parliament of Religions in Chicago became a major event in that history. Modern presentations often emphasized universal spiritual identity, the compatibility of religious paths, and the possibility of direct realization. These themes drew on older materials while reflecting new institutional, social, and intellectual circumstances.
Such presentations helped make Advaita’s identification of atman and brahman internationally familiar, but they could also obscure the plurality of Indian self-theories. “The Hindu view of the self” is frequently used for a specifically nondualist position that Rāmānuja, Madhva, or Nyāya philosophers would reject. Modern scholarship accordingly distinguishes claims made in particular Sanskrit texts, systematic interpretations developed by schools, and later public formulations of Hindu spirituality.
In comparative philosophy of mind, atman invites comparison with the thinking substance of René Descartes, the continuity accounts associated with John Locke, and David Hume’s criticism of an introspectively discoverable enduring self. The comparisons are useful only if differences remain explicit. Advaita’s self is not an individual thinking substance with a private sequence of thoughts, and Nyāya’s self is not defined by memory in precisely Locke’s manner. Hume’s discussion of perceptions and Buddhist analyses of experience likewise emerge from different argumentative and soteriological settings.
Contemporary work on personal identity and subjectivity raises related questions: whether consciousness requires a subject, whether self-awareness is prior to reflection, and whether psychological continuity needs an enduring substance. Indian discussions contribute sophisticated distinctions among the body, mental states, ego, agency, and awareness. They do not, however, constitute one ready-made answer to current scientific questions about the brain.
An especially important interpretive distinction is between the minimal subject of experience and a metaphysically permanent self. An argument that every experience has a first-person character does not by itself prove an eternal atman; conversely, criticism of a narrative personality does not automatically establish nonduality. Much comparative debate concerns precisely which transitions—from experience to subject, from subject to enduring identity, and from enduring identity to ultimate reality—can be justified.