I.

The Sāma Veda's Departure: From Recitation to Song

1.1 What the Sāma Veda Is, Precisely

The Sāma Veda is, in terms of its verbal content, almost entirely derived from the Ṛgveda: the vast majority of its approximately 1,875 verses (sāmans) are Ṛgvedic verses re-assigned to a new ritual function and subjected, in that re-assignment, to an elaborate set of melodic and vocal modifications (vikāras) that transform them, often quite dramatically, from their original recitation-format into a sung or chanted format technically distinct from any other Vedic recitation-style. It is this transformation — the application of organised melodic structure to an already-verbally-fixed text — that defines the Sāma Veda's distinctive contribution to the Vedic corpus and, this paper argues, to the Indian tradition's developing psychology of affect.

The standard academic description of the Sāma Veda as "the Veda of melodies" — accurate as far as it goes — tends to obscure the more specific and more consequential point: the Sāma Veda is the Vedic corpus' first fully systematic application of organised tonal differentiation to linguistic material, and it is simultaneously — because the linguistic material being melodically differentiated is material specifically intended to produce determinate psychological and ritual effects — the tradition's first fully systematic attempt to engineer specific feeling-states through the deliberate manipulation of tonal structure. The "melodies" are not decorative additions to a verbally complete text; they are the technical instrument of an affective-engineering project whose full scope the tradition's own commentarial literature, examined in Section III below, makes explicit.

1.2 The Technical Innovation: Vikāras as Affective Differentiation

The Sāma Veda's technical literature — preserved in the Sāmavedic Prātiśākhyas, the Puṣpasūtra, the Naradīya Śikṣā, and the later Saṃgīta-ratnākara's retrospective systematisation — describes a set of specific modifications (vikāras, literally "changes" or "transformations") applied to the underlying Ṛgvedic verse-text in the process of converting it into sung Sāman. These modifications are both linguistic-phonological (involving the elongation, repetition, insertion of syllables, and addition of meaningless syllables called stobhas — such as the ubiquitous hā, hau, hoi, o, hovā — that function, the tradition itself emphasises, not semantically but acoustically and affectively) and tonal (involving the specific pitch-contour, the sāma-melody proper, that is superimposed upon the syllabic substrate). This paper's central claim about the vikāras is that they constitute a bheda-operation applied specifically to the material of tone-and-feeling: just as Pāṇini's grammatical rules differentiate the continuous acoustic stream into discrete phonemes, the Sāmavedic vikāras differentiate the continuous pitch-space of the human voice into discrete, functionally distinct, and affectively specific tonal positions — of which the three accents (Section II) are the system's foundational minimum.

1.3 The Ṛgvedic Baseline: Why Pre-Sāman Recitation Does Not Achieve Affective Technology

To understand the Sāma Veda's departure, it is necessary to specify, with some precision, what the Ṛgvedic and Yajurvedic recitation-traditions already achieve — and what they do not. Ṛgvedic recitation employs the same three-accent system (udātta, anudātta, svarita), but its application of these accents follows the grammatical-accentual rules of Vedic Sanskrit as a natural language: the accents fall where they must for the words being recited, not where a superimposed melodic design requires them. The Yajurvedic recitation-tradition (particularly the Kṛṣṇa Yajurveda's style of prose- mantra recitation) similarly employs accent in a grammatically-determined rather than melodically-engineered fashion. What this means, in this paper's vocabulary, is that Ṛgvedic and Yajurvedic recitation apply the accent-system's bheda-operation to tone, but apply it in service of verbal precision rather than in service of affective differentiation. The accent marks where phonetic stress must fall to preserve semantic and grammatical intelligibility; it does not, in these traditions, arrange tonal differentiation in patterns specifically designed to produce, sustain, modulate, and resolve determinate feeling-states.

The Sāma Veda's departure consists precisely in decoupling the accent-system from its grammatically-determined, semantic-precision function and redeploying it — along with the vikāras' supplementary apparatus — in service of a different function: the deliberate organisation of tonal experience into feeling-shaping patterns. That this decoupling involves considerable freedom from the grammatical constraints governing spoken Sanskrit (hence the vikāras' insertions, elongations, and stobha-additions, which would be grammatically impermissible in a recited text) is not a deficiency but is precisely the technical sign of the new function being served: when tone is no longer in service of phonemic precision, it becomes available for affective engineering.

II.

The Three Accents as the Minimal Affective Bheda-System

2.1 Three Positions in Pitch-Space: Why Minimum Three?

The bheda-operation documented in Part Two at the phonemic level (Section I) and at the chromatic level (Section 2.2) always requires, at minimum, a distinction between two positions in the relevant continuum — the most minimal bheda is a binary cut, dividing the continuum into two regions. The Vedic accent-system's three positions (udātta, anudātta, svarita) rather than two reflects a specific requirement of affective differentiation that distinguishes the tonal domain from the phonemic domain: in phonemic differentiation, the relevant parameter is categorical distinctness (this is /k/, not /g/), and two positions along a binary contrast are in principle sufficient for a given contrast; in affective differentiation, however, the parameter at issue is relational movement — the felt quality of rising, falling, and the boundary-condition between them — and this relational structure requires, at minimum, a three-way distinction to be expressible: high (from which falling is possible), low (toward which falling tends), and the falling-from-high itself as a distinct, named, affectively specific third position neither simply high nor simply low.

This is not a post-hoc rationalisation of a system that happened to have three terms; the Prātiśākhyas' own analysis of the accent-system's internal logic makes clear that svarita is defined relationally — as a pitch-movement rather than a pitch-position — and that its affective specificity derives precisely from this relational character. The Taittirīya Prātiśākhya's account of svarita as the "dependent" accent, always understood in relation to the udātta it follows, establishes that the three-term system is not three independent, equally atomic categories but a structured system in which the third term is constitutively relational — the movement from the first toward the second. This structural feature will be directly relevant in Section VIII's examination of how the three-accent system expands into the seven-svara (sapta-svara) structure of classical music: the relational character of the third accent (svarita-as-movement) anticipates and, this paper argues, structurally generates the later understanding of musical intervals not as isolated pitch-positions but as movement-vectors between positions.

2.2 The Prātiśākhya Literature and the Technical Precision of Accent-Placement

The Sāmavedic Prātiśākhyas — technical manuals for the correct performance of Sāmavedic chant, of which the most important for this paper's purposes are the Ārṣeyakalpa and the various Gāna-texts specifying melodic patterns for specific Sāmans — document the accent-system with a level of technical precision that is, in this paper's context, philosophically significant: it establishes that the affective differentiation the Sāma tradition achieves is not merely a felt or intuitive phenomenon but a formally specifiable, teachable, and transmissible one. A given sāman can be learned, its melodic contour precisely specified, its accent-placements and vikāra-modifications enumerated and prescribed, such that a student trained in the tradition performs it with the same affective-engineering intention as the tradition's originators — and, the tradition claims, with the same affective-engineering effect.

This transmissibility is the crucial technical criterion distinguishing the Sāma tradition from what might otherwise be dismissed as the merely personal, idiosyncratic emotional responses that any music might produce in any listener. The claim that a given sāman reliably produces a specific affective state in a properly prepared recipient is a strong claim — and one that the Prātiśākhya literature's technical precision is itself evidence for, since that level of precision would be pointless if the relationship between melodic structure and affective response were merely contingent, culturally arbitrary, or simply a matter of individual taste. The Prātiśākhyas write as if melodic structure and affective effect are as systematically connected as phonemic structure and semantic meaning — and this paper takes that writing seriously as a datum about the tradition's own self-understanding, while Section V examines the contemporary neuroscientific evidence for whether that self-understanding is warranted.

2.3 Insufficiency of Three: Why Svarabhaktis and Vikāras Supplement the Basic System

The three-accent system, as the Prātiśākhyas make clear, is insufficient by itself for the full range of affective modulation the sāmagāna tradition achieves. The addition of svarabhaktis (vowel-colourings or transitional sounds inserted between syllables to facilitate melodic movement), vikāras (the full set of modifications already listed in Section 1.2), and the Gāna-texts' specification of multiple overlapping melodic contours (the four gānas: grāmageyagāna, āraṇyagāna, ūhagāna, and ūhyagāna, each a different melodic treatment of the same underlying sāman-text) indicates that the three-accent system is the skeleton of the affective-engineering apparatus, not the apparatus itself. The skeleton provides the minimal bheda — three positions in pitch-space, elevated from the acoustic continuum into distinct affective categories — but the flesh of the apparatus is the melodic elaboration built upon that skeleton.

III.

Sāmagāna: Chant as Affective Engineering — The Tradition's Own Self-Understanding

3.1 The Chāndogya Upaniṣad as Internal Philosophical Commentary

The Sāma Veda's most important philosophical document is not a text within the Saṃhitā itself but the Chāndogya Upaniṣad, one of the oldest and most extensively philosophical of the principal Upaniṣads, which belongs to the Sāmavedic tradition (Chāndogya = "of the Chāndoga reciters," one of the major Sāmavedic schools). The Chāndogya Upaniṣad devotes its opening chapter — and scattered passages throughout the text — to an extended philosophical reflection on what the Sāman is, what it achieves, and why its sung, melodically-structured form is not a deficient, ornamental version of a primarily verbal text but is, rather, the text's full realisation as a psychological and cosmic instrument.

Chāndogya Upaniṣad I.1 opens with the famous assertion that one should meditate on the udgītha (the specific sung portion of the Sāman chant performed by the udgātṛ-priest) as Om — and then immediately, in I.1.2–7, elaborates why: because the udgītha, being the sung form of the breath's own most fundamental vibration, carries within its melodic structure something of the prāṇa's own movement, and meditation on this structure in its sung form produces a quality of awareness — the text does not hesitate to specify — that meditation on the same verbal content in spoken form does not produce. This is, in this paper's vocabulary, the Chāndogya Upaniṣad's own theoretical claim for what melodic bheda achieves that verbal bheda does not: a specific quality of antaḥkaraṇa-state, produced by the tonal structure's action on the citta-substrate, that differs qualitatively and not merely aesthetically from what verbal-semantic content produces in the same substrate.

3.2 The Five-Fold Analysis of the Sāman

Chāndogya Upaniṣad II.2 introduces what the tradition calls the pañcavidha-sāman — the fivefold analysis of the Sāman — which divides the sung Sāman into five named parts: hiṃkāra (the preparatory intonation), prastāva (the opening section, sung by the prastotr-priest), udgītha (the main body, sung by the udgātṛ), pratihāra (the response), and nidhana (the concluding section). These five divisions are not merely structural conveniences for assigning liturgical roles to different priests; the Chāndogya Upaniṣad immediately, and with evident theoretical deliberateness, proceeds to map this five-fold structure onto various other five-fold structures in the natural world and in experience — the five breaths (prāṇa, apāna, vyāna, samāna, udāna), the five seasons, the five worlds — in a way that signals the Upaniṣad's understanding of the sāman's five-fold structure as a microcosmic instance of a universal five-fold pattern.

For this paper's purposes, the most important feature of the pañcavidha analysis is not its cosmological mappings but its implicit psychology: by identifying five distinct phases of the chant, each with a specific tonal character (the hiṃkāra's preparatory, expectant quality; the prastāva's opening, forward-reaching character; the udgītha's sustained, central, luminous character; the pratihāra's responsive, answering quality; and the nidhana's concluding, settling, resolution character), the Chāndogya Upaniṣad is implicitly specifying a temporal affective arc — a structured sequence of feeling-states produced by the chant's temporal unfolding — that is precisely what this paper means by affective technology: not the production of a single, sustained feeling-state, but the deliberate engineering of a sequence of qualitatively distinct, structured, and transitionally connected affective states.

3.3 Saṃgīta's Earliest Self-Understanding: The Nāradīya Śikṣā

The Nāradīya Śikṣā — one of the oldest surviving texts specifically concerned with musical theory as distinct from liturgical recitation-theory — marks the point at which the technical apparatus developed in the Sāmavedic tradition begins to be reflected upon as a musical-and-affective system in its own right rather than purely as a liturgical one. The Nāradīya Śikṣā's treatment of the seven svaras (tonal notes — ṣaḍja, ṛṣabha, gāndhāra, madhyama, pañcama, dhaivata, niṣāda), and its explicit account of how each svara is associated with a specific animal-cry, a specific deity, a specific emotional quality, and a specific physiological location in the body, represents the first systematic attempt — more systematic than any passage in the Chāndogya Upaniṣad — to specify the affective content of tonal differentiation: not merely that tonal differentiation produces feeling-effects, but that specific tonal positions produce specific, nameable, body-located feeling-effects.

The Nāradīya Śikṣā's seven-svara system will be examined in detail in Section VIII as the structural expansion of the three-accent minimal system. For the present section's purposes, what matters is the Nāradīya Śikṣā's explicit crossing of a threshold: the movement from liturgical to aesthetic self-understanding, from sāmagāna as a form of ritual efficacy whose psychological effects are important but secondary to its cosmic function, to sangīta as a practice whose psychological effects are themselves the primary object of systematic attention and description. This crossing — not yet the Nāṭyaśāstra's full rasa-aesthetic, but the first clear movement toward it — is the structural moment Part Three is most concerned to document.

IV.

Bhāva: Felt-Becoming as Distinct from Emotion and Sensation

4.1 The Terminological Problem

The most consequential terminological decision this paper must make — and one that Part Four's full treatment of the Nāṭyaśāstra will depend on getting right in the present paper — is how to translate, or rather how to resist translating, the Sanskrit term bhāva. The word is routinely rendered in secondary literature as "emotion," "feeling," "mental state," "mood," or some combination of these; each of these renderings captures something of what bhāva names while losing something essential, and this paper's argument requires a rendering more precise than any of the standard options.

The difficulty is etymological and philosophical simultaneously. Bhāva derives from the root √bhū, "to become," "to be," "to exist" — the same root that generates the cosmological pair bhava (existence, becoming) and abhava (non-existence). A bhāva is, in its root sense, a becoming, a dynamic process of actualisation — not a state in the sense of a static condition, but an ongoing process of being-in-a-particular-way, a continuous enactment rather than a fixed position. The Nāṭyaśāstra's technical use of bhāva — particularly in its distinction between sthāyibhāvas (enduring or standing bhāvas, the eight or nine basic feeling-orientations that correspond to the rasas) and vyabhicāribhāvas (transient or wandering bhāvas, the thirty-three subsidiary feeling-orientations that support, modify, and flow into the sthāyibhāvas) — presupposes this dynamic sense: a bhāva is not what you feel but how your citta is currently becoming, which means that bhāva is always in process, always oriented toward a next moment, always enacted in the temporal unfolding of experience rather than simply present as a static quality.

4.2 Bhāva and Citta: The Substrate Relationship

The relationship between bhāva and citta — the former being the dynamic process, the latter being the substrate in which and through which the process occurs — is the psychologically crucial relationship this paper must establish before the neuroscientific literature of Section V and the saṃskāra-analysis of Section VI can be developed. Part One's Section 5.3 established citta as the antaḥkaraṇa's memory-and-saṃskāra substrate: not a static repository but a dynamic field of tendencies, impressions, and dispositions that shapes how the other antaḥkaraṇa-functions (manas, buddhi, ahaṃkāra) operate by providing the background context — the accumulated history of prior experience — within which each new experience occurs. Bhāva, on this account, is what happens in citta when a specific pattern of input — a melodic pattern, a dramatic enactment, a visual form, a verbal formulation — activates a specific set of citta's accumulated tendencies, bringing them into dynamic, felt expression in the present moment.

This account of bhāva-as-citta-activation is what distinguishes it from both "emotion" in the modern psychological sense (which typically refers to a complex event involving cognitive appraisal, physiological arousal, action-readiness, and subjective feeling, but does not necessarily involve the citta-substrate's accumulated saṃskāra-history in the specific way bhāva does) and from "sensation" (which refers to immediate, substrate- independent, present-moment hedonic valence without the specifically dispositional, history-shaped character bhāva has). A bhāva is felt — it has hedonic and experiential quality — but it is felt as a becoming that is simultaneously a remembering: each bhāva activates not merely a present-moment affective quality but the entire citta-history of prior instances of that quality, which is why the same melodic phrase can produce a progressively richer and more layered affective response in a listener whose citta has been more extensively shaped by prior musical experience than in one whose has not.

4.3 The Eight Sthāyibhāvas: Bhāva's Fundamental Taxonomy

The Nāṭyaśāstra's eight sthāyibhāvas — rati (love-longing), hāsa (mirth), śoka (grief), krodha (anger), utsāha (heroic energy), bhaya (fear), jugupsā (disgust), and vismaya (wonder/astonishment), to which the later tradition adds śama (tranquillity) as a ninth — will receive their full treatment in Part Four as the affective architecture underlying the eight (or nine) rasas. Part Three's task is more limited but necessary: to establish that these eight categories do not emerge ex nihilo with Bharata's Nāṭyaśāstra but are the outcome of a prior differentiation process — a bheda-operation applied to the affective field — that the Sāma Veda's melodic technology begins, even if the Sāma tradition itself never arrives at the Nāṭyaśāstra's systematic nomenclature and theoretical architecture.

The evidence for this claim is partly structural and partly historical. Structurally: the Sāma Veda's chant-tradition already distinguishes, in its practical organisation of different Sāmans for different ritual occasions and purposes, between melodic forms associated with śoka-like (grief-adjacent) affective states (the Sāmavedic funeral Sāmans, with their characteristic descending melodic movement and predominance of anudātta), those associated with utsāha-like (heroic, energising) affective states (the battle-associated Sāmans with their udātta-dominant, ascending, rhythmically forceful character), those associated with rati-like (longing, yearning) affective states (the Keśī-Sāman and related forms, with their sinuous, extended melodic movement and heavy svarabhakti-elaboration), and so forth. The classificatory work of distinguishing these different affective registers — the practical taxonomy preceding the theoretical one — is already begun in the Sāma tradition's organisation of its own material. Historically: the Nāṭyaśāstra's rasa-taxonomy does not descend directly from the Sāmavedic chant-tradition alone; it synthesises multiple prior traditions including the Vedic, the epic-bardic, and the early dramatic. But the Sāma tradition's melodic differentiation of the affective field is among the earliest and most technically precise of the inputs into this synthesis.

V.

The Neuroscience of Melodic Affect

5.1 Why This Section Is Necessary and Where It Fits

Part Two established the methodological precedent for this paper's engagement with contemporary empirical science: the categorical perception literature was examined not as a source of validation for ancient Indian insights (a form of reverse-chronological argument that would be methodologically unsatisfying) but as an independently arrived-at, formally rigorous documentation of phenomena the Vedic and grammatical traditions had also documented, using different methods, at different levels of precision, in service of different goals. The same methodology applies in the present section: the neuroscientific literature on melodic affect — a research area that has expanded considerably in the decades since Stefan Koelsch's 2014 landmark review and that now encompasses neuroimaging, psychophysiological, electrophysiological, and pharmacological approaches to music's emotional effects — is examined here not to "validate" the Sāma Veda's claims but to provide a contemporary empirical anchor for the specific psychological mechanisms this paper has proposed are at work in sāmagāna's affective-engineering function.

5.2 Frisson and the Physiology of Chills: Expectation-Violation as Affective Bheda

One of the most robustly documented phenomena in the music-emotion literature is frisson — the psychophysiological response of cutaneous chills or goosebumps (piloerection) and a characteristic shivering sensation produced by specific moments in musical performance, particularly moments involving unexpected harmonic movement, the entry of a new voice or instrument at a surprising register, or the resolution of prolonged harmonic tension. Frisson is documented to involve activation of the nucleus accumbens and the release of dopamine — the same neurochemical system involved in reward-processing more generally — and its occurrence is predicted not primarily by the overall emotional valence of the music but specifically by expectation-violation: the moment at which the musical structure violates a prediction generated by the listener's prior musical experience, and then, characteristically, resolves that violation in a way that is simultaneously surprising and, in retrospect, felt as inevitable.

This expectation-violation-then-resolution structure is, in this paper's framework, a direct instantiation of the bheda-operation at the affective level: the listener's prior musical saṃskāra generates an expectation (a predicted future tonal position in the melodic continuum), the music's actual movement constitutes a discrimination (the actual path taken differs from the predicted path, creating a boundary-event — a moment of affective bheda, the experience of a cut in the expected tonal fabric), and the resolution that follows — the music's arrival at a position that, despite violating the prediction, is immediately felt as right — constitutes the affective equivalent of Stage 3's discrete units: the new, post-bheda affective position, now fully available to feeling-awareness and to citta's saṃskāra-formation, in a way that a merely predicted, non-violated tonal movement would not have achieved.

5.3 The ARAS System and Arousal Modulation

A second empirically well-established mechanism in the music-affect literature concerns the ascending reticular activating system (ARAS) and its involvement in musically-induced changes in arousal level. The ARAS — the brainstem network responsible for regulating the overall arousal level of the cortex, and hence the general readiness of the entire antaḥkaraṇa-complex for attentive, responsive processing — is known to respond to specific acoustic parameters including tempo, rhythmic regularity, loudness, and pitch-register. Music with high tempo, strong rhythmic regularity, high loudness, and high pitch-register tends to produce ARAS activation (increased arousal), while music with slow tempo, irregular rhythm, low loudness, and low pitch-register tends to produce ARAS deactivation (decreased arousal, relaxation). This mapping — standard parameters of melodic structure onto standard parameters of arousal-state — provides the neurobiological mechanism for exactly the gross-level arousal effects the Sāma Veda's different Sāman- categories achieve: the battle-associated Sāmans' udātta-dominant, rhythmically forceful, ascending character maps directly onto the ARAS-activating end of the spectrum; the funeral Sāmans' anudātta-dominant, slow, descending character maps onto the ARAS-deactivating end.

What the neuroscientific literature adds to what the Sāma tradition already knew at the level of practical experience is the mechanism-specification: ARAS involvement tells us why the effects the Sāma tradition documented are reliable and not merely idiosyncratic, why they are transmissible across different performers and recipients (because the acoustic parameters that modulate ARAS are relatively invariant across individuals within a shared musical culture), and why they are trainable and improvable through repeated exposure (because ARAS-modulation through music becomes more precise and more rapid as the citta-substrate accumulates the musical saṃskāras that enable finer expectation-generation and finer expectation-violation detection).

5.4 The Default Mode Network and Musical Absorption

A third neuroscientific finding directly relevant to this paper's argument concerns the default mode network (DMN) — the network of cortical regions (medial prefrontal cortex, posterior cingulate cortex, angular gyrus, and related areas) that is characteristically more active during rest, internally-directed thought, and autobiographical memory-processing than during externally-focused task performance. Research on musical absorption — the phenomenologically distinctive state of "being carried" by music, in which self-referential awareness temporarily diminishes and the musical experience occupies a larger proportion of awareness — has documented that deep musical absorption involves characteristic DMN modulation: the DMN's self-referential processing is simultaneously reduced (the listener is less aware of themselves as an observer distinct from the music) and integrated with the musical processing networks in a specific way that supports what is subjectively experienced as the "loss of self" in music.

This DMN-modulation finding is directly relevant to the Chāndogya Upaniṣad's philosophical claims about sāmagāna, and to this paper's own account of what the Sāma tradition is achieving at the deepest level of its affective-engineering function. The Chāndogya Upaniṣad's insistence that meditation on the udgītha produces a form of awareness qualitatively different from, and not achievable through, the same verbal content in spoken form (Section 3.1 above) is precisely a claim about DMN-modulation in the language available to the tradition: the sung form of the Sāman produces a degree of self-referential absorption-and-dissolution that the spoken form does not, because the melodic structure's sustained, total engagement of the processing system achieves the DMN-modulation that verbal-semantic content alone does not trigger. The specific Upaniṣadic recommendation to meditate on the udgītha as Om is, on this reading, a practical instruction for achieving this DMN-modulated state through a specific technique — the identification of the melodic practice with the prāṇic continuum it is heard as expressing — that functions, in neurobiological terms, to deepen and sustain the DMN-modulation that the melodic practice alone initiates.

VI.

Citta, Saṃskāra, and the Psychology of Affective Memory

6.1 How Repeated Melodic Exposure Shapes the Citta-Substrate

The relationship between citta and saṃskāra, established in Part One's Section 5.3 and revisited in Part Two's Section 4.3 in connection with the saṃskāra-conditioning of category-boundary placement, now receives its fullest development in the present paper: the Sāma tradition's mechanism of affective education is precisely the accumulation of melodic-affective saṃskāras in the citta-substrate through lifelong, highly structured, highly repetitive exposure to the Sāmavedic melodic repertoire. The Vedic educational system (gurukula) for the Sāmavedic priest involved years of daily exposure to the Sāmavedic chant-repertoire — not merely passive listening but active embodied participation, since the student's own vocal apparatus was the instrument through which the melodic patterns were internalised. This active, embodied, vocal participation means that the saṃskāras being deposited in the citta-substrate were not merely auditory-perceptual impressions but kinesthetic-motor-vocal impressions: the felt sense of producing these tonal movements with one's own body, the proprioceptive and interoceptive dimensions of the tonal contour, were part of what was being impressed into the citta alongside the auditory qualities.

6.2 Saṃskāra as Affective Groove: The Psychology of Musical Training

The concept of saṃskāra — literally an "impression," "perfection," or "refinement" produced by repeated action, the Sanskrit term standardly translated as "latent impression" or "subconscious tendency" in secondary literature — is, in the context of affective experience specifically, best understood as what might be called an affective groove: a track worn into the citta-substrate by repeated affective experience that makes future instances of similar experience more readily available, more precisely defined, and more efficiently recognisable. Just as a physical groove in a surface channels future movement along its already-worn track, a citta-saṃskāra channels future affective responses toward the affective forms that prior experience has already differentiated, named (if only implicitly), and stabilised.

The contemporary music-psychology literature provides empirical evidence for this groove-formation process in terms of what it calls musical expertise effects: listeners with extensive formal musical training show measurably different patterns of emotional response to music than non-trained listeners — not simply more intense responses, but more precisely differentiated responses, with finer-grained discrimination between affective nuances (the trained listener's ability to distinguish, and to feel as distinct, affective qualities that the untrained listener experiences as a single, undifferentiated emotional response). This expertise-driven differentiation is exactly what the saṃskāra- model predicts: the accumulated impressions of prior, structured melodic experience in the citta-substrate enable finer-grained discrimination (a more precisely placed Stage 2 operation) when new melodic experience is encountered, producing more finely differentiated Stage 3 affective units, and therefore a richer, more nuanced, and more specifically cultivatable affective life.

6.3 The Generational Logic: Why Lifelong Immersion Matters

The Sāma tradition's approach to affective education is not a quick technique for producing immediate emotional effects; it is, structurally, a multigenerational project of citta-refinement achieved through lifelong melodic immersion. The gurukula's demand that a student spend years — often a decade or more — in active daily practice of the Sāmavedic repertoire before being considered competent to perform the chants in a ritual context reflects the tradition's implicit understanding that the saṃskāra-substrate requires sustained, graduated, embodied formation before it can support the level of affective precision the ritual context requires. The Sāman performed by a priest whose citta has been shaped by decades of active melodic practice is, on this account, not merely technically more accurate but affectively more precise — capable of producing, through the combination of tonal skill and citta-refinement, the specific, targeted affective effects the ritual requires, in a way that the same notes performed by an insufficiently trained performer could not produce.

This generational, accumulatively citta-shaping logic distinguishes the Sāma tradition's affective technology from any quick technique or shortcut to emotional experience, and it is the specific feature of the tradition that Part Four's rasa-aesthetic — with its own emphasis on the trained spectator (sahṛdaya) whose prior aesthetic experience has prepared the citta for rasa-reception — will inherit and systematise into a general theory of aesthetic competence. The sahṛdaya is, in Part Four's terms, the person whose citta has been sufficiently shaped by prior aesthetic-affective saṃskāras to be capable of the full rasa-experience that the Nāṭyaśāstra describes and prescribes — and the direct ancestor of the sahṛdaya, in terms of the historical and structural development this series traces, is the Sāmavedic priest whose citta has been shaped by lifelong melodic immersion into the precise affective sensitivity the Sāman requires.

VII.

From Liturgy to Cultivation: The Deliberate Turn

7.1 The Institutional Threshold

There is a specific historical and structural moment at which the affective-engineering function of melodic practice becomes the primary object of reflection rather than a secondary consequence of a primarily liturgical or ritual function. Locating this moment precisely is not straightforward — the evidence is literary and textual, not archaeological, and the texts in question were accumulated over centuries rather than composed at a single identifiable moment — but the convergence of evidence from several directions makes it possible to characterise the threshold even if it cannot be precisely dated. The threshold is the moment at which the question "what melodic structure produces this affective state in this recipient?" begins to be asked independently of the question "what melodic structure is ritually prescribed for this liturgical occasion?" — that is, the moment at which affective efficacy becomes a criterion of evaluation in its own right rather than merely an instrument of ritual efficacy.

This threshold is crossed — or rather, begins to be approached — in the earliest strata of two textual traditions: the Chāndogya Upaniṣad's philosophical reflection on sāmagāna (Section III), which raises the question of what the sung form achieves that the spoken form does not — a question that presupposes the affective dimension of musical experience as a legitimate object of philosophical inquiry; and the Nāradīya Śikṣā's svara-taxonomy (Section 3.3), which assigns specific affective content to specific tonal positions in terms that go considerably beyond what liturgical prescription requires, indicating that the affective dimension of tonal differentiation is being attended to for its own sake.

7.2 Gandharva-Vidyā: Music as Distinct Knowledge-Domain

The earliest Sanskrit term for the deliberate study of music as a distinct domain of knowledge — as distinct from the liturgical study of the Sāmavedic chant as a component of Vedic education — is gandharva-vidyā, knowledge of the gandharvas, the celestial musicians of the Sanskrit cosmology who are held to be the original masters of the musical art. The term occurs in the Chāndogya Upaniṣad (VII.1.2) in Nārada's inventory of what he knows, and its appearance in that context — a list of knowledge-domains alongside ṛk, sāman, yajus, itihāsa-purāṇa, and others — establishes that by the Upaniṣadic period, music was already recognised as a distinct knowledge-domain that could be the object of systematic study.

The significance of gandharva-vidyā's emergence as a named, institutionally distinct knowledge-domain is that it marks the institutionalisation of the deliberate turn already identified in Section 7.1: when music is studied as gandharva-vidyā rather than as a component of Sāmavedic liturgical training, the question being asked is explicitly about the knowledge of music's properties, capacities, and effects — including its affective effects — rather than exclusively about its correct liturgical performance. Gandharva-vidyā is, in this sense, the institutional recognition that affective cultivation through tonal practice is a legitimate and teachable skill, not merely a byproduct of liturgical competence.

7.3 The Transition to Dramatic Context: Sāmagāna and the Origins of Nāṭya

The Nāṭyaśāstra's own account of its origins — in the famous Nāṭyaśāstra I.9–15, the Brahma-creation narrative in which Bharata receives the science of dramatic performance from Brahma who has compiled it from the four Vedas (words from the Ṛgveda, song from the Sāmaveda, abhinaya from the Yajurveda, rasa from the Atharvaveda) — explicitly includes the Sāmavedic melodic tradition as one of the Nāṭyaśāstra's constitutive sources. The claim that rasa derives from the Atharvaveda (rather than from the Sāmaveda) is notable and will require examination in Part Four; for the present paper's purposes, the important point is that song — specifically Sāmavedic-tradition song — is assigned a specific and irreplaceable function in the Nāṭyaśāstra's synthesis: it is the vehicle through which the dramatic performance achieves its affective modulation of the audience's citta-substrate, and this function is understood to derive specifically from the Sāmavedic tradition's developed capacity for melodic affective-engineering, not from anything the dramatic context itself provides independently.

VIII.

Rāga's Origin in the Sāma Accent-System: The Internal Proliferation of Bheda

8.1 From Three Accents to Seven Svaras: Structural Necessity

Section 2.3 established that the three-accent minimal system (udātta, anudātta, svarita), supplemented by the vikāra-apparatus, is insufficient by itself for the full range of affective modulation the sāmagāna tradition achieves. The expansion from the three-accent system to the seven-svara (sapta-svara) system — ṣaḍja (sa), ṛṣabha (ri), gāndhāra (ga), madhyama (ma), pañcama (pa), dhaivata (dha), niṣāda (ni) — is a case of bheda's own internal proliferation, exactly analogous in structural logic to the expansion from the Māheśvara sūtras' phonemic categories to Pāṇini's full phonological theory: the more finely one requires the discrimination of affective states, the more finely one must differentiate the tonal system that mediates those states.

The seven-svara system is first documented in a form directly traceable to Sāmavedic practice in the Nāradīya Śikṣā, which assigns each svara a specific association — with the animal-cry by which it is supposedly imitated (sa = peacock, ri = bull, ga = goat, ma = crane, pa = cuckoo, dha = horse, ni = elephant), with a specific grāma (tonal scale in a particular register), and with a specific deity. What matters for this paper's argument is not the cosmological associations (though these reflect the tradition's understanding of the svaras as cosmologically, not merely acoustically, significant) but the specific affective-emotional associations each svara carries — associations documented in the Nāradīya Śikṣā and substantially elaborated in the later saṃgīta-śāstra literature including the Bharata's Nāṭyaśāstra (Chapter XXVIII), the Mataṅga's Bṛhaddeśī, and the Saṃgīta-ratnākara.

8.2 The Grāma System and Affective Modal Differentiation

The sapta-svara's affective differentiation does not operate through individual svaras in isolation but through the grāma — the modal scale, the specific arrangement of the seven svaras into a particular intervallic structure — and subsequently through the rāga, the melodic archetype that specifies which svaras are used, in what order, with what emphasis, and with what ornamental elaboration in a given composition or improvisation. The grāma-system (the Nāṭyaśāstra recognises three primary grāmas: ṣaḍjagrāma, madhyamagrāma, and gāndhāragrāma, of which the last has largely fallen out of use in the historical tradition) and the rāga-system (in which hundreds to thousands of specific melodic archetypes are identified, named, and assigned specific affective characters — rāga-bhāvas — in the classical tradition) represent the full, systematic elaboration of what began as the Sāma Veda's three-accent minimal affective-bheda system.

The rāga's affective-engineering capacity — its ability to reliably produce specific, named, nameable feeling-states in a properly prepared listener — is, on this paper's account, the mature fruit of a developmental process that began with the Sāma Veda's first systematic application of tonal differentiation to the affective field. The rāga is what affective bheda looks like when it has been refined, over centuries of musical and contemplative practice, to a level of precision that allows the melodic structure to specify not merely a gross affective register (grief-adjacent versus heroic-adjacent) but a precisely individuated affective character — the specific emotional-aesthetic quality of Bhairavi at dawn, which is different from the specific emotional-aesthetic quality of Todi at the same time, which is different again from Bhairav, each rāga carrying its own irreducible affective signature that no other rāga can replicate despite sharing some of the same svaras.

8.3 The Bhāvarāga Relationship: Affect Produces Form

A feature of the classical Indian music-theory tradition that is directly relevant to this paper's argument — and that distinguishes the Indian understanding of the melody-affect relationship from the more common Western musicological understanding — is the tradition's insistence that the relationship between rāga and bhāva runs in both directions: not only does a given rāga reliably produce a specific bhāva in a properly prepared listener, but a given bhāva — a specific felt-becoming in the practitioner's own citta — is said to give rise to and inform the melodic choices that constitute the rāga in performance. The practitioner singing Bhairavi is not merely applying a melodic formula that happens to produce a Bhairavi-bhāva in the audience; the practitioner's own Bhairavi-bhāva — the specific state of the citta that Bhairavi names — informs, from the inside, which svaras are emphasised, how the gamakas (ornamental pitch-slides) are shaped, what the tempo and rhythmic feel are, in a way that cannot be fully captured by any external description of the rāga's formal properties alone.

This bidirectional relationship — rāga produces bhāva, bhāva shapes rāga — is, in this paper's framework, a direct instance of the more general relationship between bheda-operation and its affective result: the discrimination is not a one-directional imposition of form on formless content, but a mutual shaping in which the content's own tendencies (the citta's saṃskāra-shaped disposition toward certain affective forms) partly determine how the discrimination falls, while the discrimination's result (the specific affective form produced) in turn shapes the citta that will perform future discriminations. This circularity is not a methodological defect but is precisely the form affective education takes: each practice session both uses and shapes the citta, such that the cumulative result of sustained practice is a citta progressively more sensitively and more precisely oriented toward the affective forms the practice cultivates.

IX.

Limits and the Forward Problem

9.1 What the Sāma Tradition Does Not Achieve: Three Acknowledged Limits

Methodological integrity requires this paper to specify, with the same precision it has applied to the Sāma tradition's achievements, what the Sāma tradition does not achieve — the three most important respects in which affective differentiation remains incomplete or insufficiently systematic in the Sāma Veda's own technical apparatus, and which therefore constitute genuine rather than merely apparent reasons for Part Four's treatment of the Nāṭyaśāstra to be necessary rather than redundant.

The first and most fundamental limit is the absence of a systematic theory of affective categories: the Sāma tradition differentiates the affective field practically (different Sāmans for different ritual-affective occasions) but does not provide a theoretical taxonomy of the affective field's fundamental categories. The eight (or nine) sthāyibhāvas, their mutual relationships, their hierarchy, the principle by which they constitute a complete rather than a merely arbitrary set — these are not available in the Sāma tradition's technical literature. They are the specific theoretical contribution of the Nāṭyaśāstra, and Part Four's task is precisely to develop this contribution at the level of detail it requires.

The second limit is the absence of a theory of the recipient: the Sāma tradition specifies what melodic structures produce what affective effects, and it specifies the educational process through which the performer must be trained, but it does not develop a theory of the aesthetic recipient — the person on the receiving end of the melodic engineering — as a distinct theoretical object. The Nāṭyaśāstra's concept of the sahṛdaya (the sympathetically-hearted spectator-recipient) and its theory of how the sahṛdaya's citta-constitution determines their capacity for rasa-experience is a specifically Nāṭyaśāstra-level theoretical innovation, not anticipated in the Sāma tradition, and Part Four will develop it accordingly.

The third limit is the Sāma tradition's restriction to purely melodic (acoustic, tonal) means of affective engineering. The Nāṭyaśāstra's synthesis is considerably broader: it brings together the melodic tradition, the verbal tradition (kāvya, poetic expression), the visual and gestural tradition (abhinaya, the fourfold system of expressive means), and the dramatic tradition (nāṭaka, theatrical performance) into a single integrated affective-engineering system that achieves, through their mutual reinforcement, a level of precision and reliability in producing specific affective effects that no single tradition could achieve alone. Part Four's task is to examine this synthetic integration and the theoretical architecture — rasa-theory — that makes it possible and describes it.

9.2 The Specific Advance That Justifies Part Four

The single most important specific advance the Nāṭyaśāstra makes over what the Sāma tradition provides is the concept of rasāsvādana — the "tasting" or "relishing" of rasa — which frames the affective experience the Nāṭyaśāstra cultivates not as a produced state that happens in the recipient but as an active achievement, a cognitive-affective event in which the recipient's own citta collaborates with the performance to produce an experience that transcends ordinary emotional experience by being simultaneously deeply felt and aesthetically distanced, personally resonant and universally representative. This concept — the basis of abhinavagupta's later, definitive philosophical treatment in the Abhinavabhāratī — has no real analogue in the Sāma tradition, where the relationship between chant and citta is understood as a more direct, less mediated engineering of specific states. The mediation — the specific quality of aesthetic distance and aesthetic generalisation that rasāsvādana introduces — is the Nāṭyaśāstra's most consequential philosophical innovation, and Part Four's analysis will centre on it.

X.

Forward to Part Four: From Affect to Rasa

10.1 What This Paper Has Established

The present paper has developed ten results. First, it has located the Sāma Veda's specific contribution to the history of psychology in the decoupling of the tonal accent- system from its grammatically-determined function and its redeployment as an affective- engineering apparatus. Second, it has analysed the three-accent minimal system as the affective equivalent of the phonemic minimal system — the minimum number of distinct tonal positions required to structure the pitch-continuum into discrete affective categories. Third, it has examined the Chāndogya Upaniṣad's philosophical reflection on sāmagāna as the tradition's own internal evidence for what it takes itself to be achieving: the engineering of specific antaḥkaraṇa-states, not merely the honouring of cosmic structures through melodic performance. Fourth, it has distinguished bhāva-as- felt-becoming from both "emotion" and "sensation," establishing the category that Part Four's rasa-theory will presuppose. Fifth, it has engaged the contemporary neuroscientific literature on melodic affect as an empirical anchor, identifying three mechanisms (frisson, ARAS modulation, DMN modulation) that provide a mechanism-level account of what the Sāma tradition's practical experience and reflective commentary document at the experiential and theoretical levels. Sixth, it has developed the citta-saṃskāra model of affective education as the psychological mechanism by which the Sāma tradition's lifelong melodic immersion produces the durable, transmissible, refinable affective sensitivity that constitutes a fully trained aesthetic citta. Seventh, it has traced the historical threshold at which melodic affective-engineering becomes an object of reflection in its own right — the emergence of gandharva-vidyā. Eighth, it has examined the structural expansion from the three-accent minimal system to the seven-svara and rāga systems as a case of bheda's own internal proliferation under the pressure of affective refinement. Ninth, it has specified three limits of the Sāma tradition's achievement that genuinely require Part Four to be a new contribution rather than a recapitulation. Tenth, it has identified rasāsvādana — the specifically Nāṭyaśāstra concept of aesthetic tasting — as the single most important conceptual innovation distinguishing Part Four's domain from Part Three's.

10.2 The Handoff to Part Four

Part Four — Nāṭyaśāstra I: Rasa and the Architecture of Emotion — will examine the Nāṭyaśāstra's rasa-theory as the third major elaboration of the bheda-operation this series has been tracing since Part Two: bheda applied to the affective field, now in the context of a fully synthetic, multi-modal, theoretically systematic aesthetic practice. Three contributions from the present paper feed directly into Part Four's argument. First, the bhāva-concept developed in Section IV — specifically the distinction between sthāyibhāva (enduring affective orientation) and vyabhicāribhāva (transient modifier) — provides the affective architecture on which rasa-theory builds. Second, the citta-saṃskāra model of affective education developed in Section VI provides the psychological mechanism through which Part Four's sahṛdaya-concept will be theorised: the sahṛdaya is, specifically, a person whose citta has been sufficiently formed by affective saṃskāras to be capable of the full rasa-experience. Third, the acknowledged limit identified in Section 9.2 — the Sāma tradition's relative absence of a theory of affective distancing and aesthetic generalisation — defines the specific theoretical space that rasa-theory's concept of rasāsvādana is developed to fill.

A phoneme divides sound into meaning. A Sāma Veda accent divides sound into feeling. A rāga divides feeling into specific, named, cultivatable aesthetic forms. And a rasa — this series will argue in Part Four — divides aesthetic experience into universally recognisable, impersonally transmitted, personally resonant forms of human being. The operation is the same. The material, and the stakes, change at every step. Series B · Editorial Framework