Go to Semiotic Review of Books Home Page
Go to SRB Highlights
Go to SRB Archives


SRB Archives

This editorial appeared in Volume 4 (2) of The Semiotic Review of Books.

Editorial: Sounds for Thought

by Fabio Rambelli

What is a mantra? This term refers to a loose set of peculiar Indian linguistic entities. Some are endowed with meaning in ordinary Sanskrit language, but in general they are charged only with initiatory value. Mantra is essentially a sound, a sequence of sounds functioning as an evocator of energy. According to Indian Tantric doctrines, the vibrations constituting the universe manifest themselves as linguistic sounds, as "seed syllables" (bija) which combine to form Mantras. It is virtually impossible to give a simple and exhaustive definition of Mantra: the term refers to an amazingly wide ambit of linguistic entities of various kinds, used in very different contexts, apparently shading only an Indian matrix and the fact that they amount to a non-ordinary usage of language. The most comprehensive definition remains Harvey Alper's: "a mantra is whatever anyone in a position to know calls a mantra" (Alper 1989: 4).

The concept of mantra is closely attached to Indian culture where it originally developed and in which it takes an exceptional importance. Such strong cultural characterization did not prevent mantras from spreading throughout Asia with Hinduism and Buddhism. Indian original mantric doctrines interacted on many levels with the different cultures which they penetrated, also contributing to changes in their linguistic ideas. Mantras themselves, however, remained almost unchanged; their power rests in their sound, and "in an effort to duplicate and thereby duplicate the sound" of the voice of deities proclaiming these sacred expressions (Lopez: 359) their pronunciation has been transliterated in numerous writing systems, but never forgotten.

The difficulty of defining mantric phenomena also causes problems in understanding their theoretical status. Scholarly debates concern mainly mantras' linguisticity -- i.e., whether mantras are language or not. Some scholars deny such a possibility, others explain them as speech acts, or as language games, or even as instruments of magic. In any case, all agree on the importance of the context, of the actual situation in which the usage of mantras occurs.

In Indian treatises, mantras, peculiar linguistic objects outside of the rules and conventions regulating ordinary language, never seem to take on the traits of a real independent language. The Tantric tradition, on the contrary, especially in East Asia, explicitly attributes to mantras the status of absolute language. Be that as it may, it is not easy to consider mantras as a form or a particular usage of language, conforming to linguistic rules. According to the most authoritative critical voice, Frits Staal, mantras are pieces of texts basically devoid of meaning which take on the function of ritual objects. Endowed with phonological and pragmatic properties but devoid of syntax and semantrics, they do not conform to Western and (non-esoteric) Indian theories of language, therefore they cannot be considered as either linguistic entities or speech acts. According to Staal, mantras are ritual elements, and mantric practice, as a rite, is a behaviour defined by rules but lacking with meaning and finalities, determined as it is by obscure biological constraints.

Staal thinks that meaning is an exclusive property of ordinary language denotatively used; ritual, and mantras as ritual objects, are devoid of it: "like rocks or trees, ritual acts and sounds may be provided with meaning, but they do not require meaning and do not exist for meaning's sake" (Staal 1986: p. 218). There is, however, an authoritative approach which is diametrically opposed to Staal's and according to which everything which is part of a culture must have a sense, be explicable, interpretable. An easy objection to such a position is that the listeners (included many officiants) of mantfic expressions almost never understand their meaning. Now, Wittgenstein stated that the meaning of linguistic expressions is their usage, and therefore it is not necessary to understand the meaning of an expression in order to use it correctly. In this regard, it is important to underline the fact that many esoteric mantric texts presuppose a model reader who is able to understand the meaning of mantras (see also Tambiah (1985:21)).

In order to escape the impasse to which discussions on mantras' linguisticity lead, the reformulation of the question in semiotic terms could be useful. Since semiotics deals with the abstract structure of signification systems (including, but not exclusively, verbal language), as well as the processes whereby users apply rules of these systems in order to communicate, criticize and modify the structure of signification systems themselves, it is possible to show that mantric expressions are not just mere semiotic devices, but constitute a semiotics in Hjelmsley sense. Within the semiotic system of esoteric Buddhism, for instance, mantric sign relations are constituted, in Hjelmslevian terms, by two main levels of expression (phonological and graphic, each organism in a form and a substance, and by a plane of the contents, in turn articulated on many levels but structured in form and substance of the contents.

Phonology: Mantras' phonology concerns a whole range of norms, transmitted from master to disciple, determining for each linguistic entity not only the pronounciation, intensity of voice, rhythm and melodic structure of the utterance, but also breathing, bodily posture, appropriate ritual. These rules are such that Frits Staal proposes to take into account "the importance of musical categories for explaining some of the characteristics that distinguish mantras from language" while aware of the fact that "mantras cannot be understood unless their musical character is taken into account" (Staal 1989: 65). It is hard to deny, though, that music explains only one aspect of mantras.

Syntax: It is not clear if a syntax of mantras exists. Scholars hold various positions: Frits Staal denies such a possibility, while Donald Lopez and Stanley Tambiah are ready to admit it. Certainly Buddhist mantras constitute a repertoire of fixed expressions which has not changed for centuries. Perhaps a taxonomic, distributive and componential study of mantric formulas could help to ascertain if structural and distributive constants correspond also to functional and semantric ones.

Mantric linguistic space of Tantric Buddhism is generally marked by formulas such as om at the beginning and hum or svaha at the end, between which are inserted linguistic "seeds" (bija) of deities, invocations and concepts more or less connected by a single theme. Mantras are a sort of "scrap-book language", a series of concepts and elements juxtaposed and/or superimposed. Mantric syntax consists of a set of rules establishing a sacred linguistic space through the usage of peculiar expressions, as well as a non-specified code associating by analogy a sequence of sounds to a meditative or salvational process, identifying them, through the mediation of content. In this manner, it is possible to create new mantras on the basis of a set of doctrines, deities, and aims.

Semantrics: Frits Staal drastically denies that mantras have any meaning whatsoever. In the case of East Asian mantric expressions, at least, each expression is endowed with a definite, if complex, meaning. In mantras are found many cases of conformity, representing the substantial identity of the elements concurring in constituting the mantra as a sign relation. In rigorously Hjelmslevian terms, this is sufficient to deny their linguisticity. Nevertheless, the denial of linguisticity does not affect their interpretability and therefore their nature as a semiotic system. Moreover, it is evident that the sense of mantras (at least of many among them) does not depend just on their intention (i.e.even their numerous levels of meaning) but also and above all on esoteric isotopes. Thus, mantras become the inscription, the linguistic substance of a process, a ritual, a technique, and this fact has a direct esoteric relevance.

Pragmatics: Mantric expressions are not used in everyday communicational interaction; their ambit is ritual meditation, magic. Mantras are impediments, on the one hand, for production, conservation and transmission of a certain type of knowledge; on the other hand, they assist in the transformation of such knowledge into power over reality through illocutionary and performative acts. Stanley Tambiah was perhaps the first to underline this important fact in his studies on power formulas used by Thai Buddhism.

Mantras are objects which certain cultures consider as endowed with meaning; they are used in ritual performances and allow communication between beings living on different levels of consciousness and reality. To paraphrase Harvey Alper mantras are "machines" producing particular states of consciousness, transformation of knowledge, different images of reality. Their use presupposes and, at the same time, grounds an episteme, a particular cultural relation with signs.

At the same time when mantras concur to create such a sacred communicational space, they are used in ritual to communicate with dieties, playing an illocutionary function. The use of mantras determines precise effects, such as the transformation of the world. Such a transformation is not just incorporeal, but dramatically bodily and material, since it concerns healing, worldly benefits, rebirth, becoming Buddha. Accordingly, repetition (vocal or just mental) of mantras is not only a saying or a thinking, but an actual activity producing effects upon reality. Such a conception is perfectly in accordance to the traditional Indian approach to language as instrument to do something (Comolli 1993).

Speech act theory deals with such an important active aspect of language, and many scholars have applied it to mantric phenomena, although there are numerous heuristic problems -- understandably, since speech act theory explains usages of ordinary language which have almost nothing in common with the recitation of mantras. Some objections to the application of speech act theory to mantric expressions have been raised by Frits Staal (1989). But Staal does not take into account the epistemic presuppositions of mantra usage (Rambelli 1989, 1992).

An integral application of speech acts theory to the ritual usage of language in Buddhism must also take into account phenomena not considered by Austin's original theory, i.e., the existence of "transitive" speech acts whose effects are transferred to another person, as well as "graphological" acts, language acts which are not speech acts, consisting in the transcription of written formulas in order to "illocutionarily" change reality.

Thus, it is possible and heuristically profitable to consider mantras as illocutionary expressions endowed with a particular perlocutionary function, by a value investment (faith) by the utterer and, above all, by the structure of mantric expressions themselves, motivated signs in which signifier and signified, language and reality are closely related. According to Wade Wheelock's definition, mantric speech acts are "ritual" speech acts, presupposing a different idea of language allowing non-ordinary linguistic practices.

To get a more complete grasp of the mantric phenomenon, at least as far as certain Asian traditions of thought are concerned, it is also necessary to take into account another usually neglected aspect: the writing of mantras. At the beginning writing, in accordance with mainstream Buddhist ideas of language, was considered as a mere device to convey the meaning of Buddha's words. This attitude changed around the seventh or eighth century with the development of a systematic form of Tantric Buddhism for which the sound of esoteric expressions was no longer sufficient: rituals also needed the original graphemes. In China and especially in Japan the characters of the Indian writing called siddham were used in various ways. Siddham refers to a gupta type of the brahmi writing system, used in India between the Fourth and the Eighth century, and now extant only in Japan. In this manner, siddham, earlier known only to the few translator monks, became an extremely important subject of study and practice for the Buddhist monks. The East Asian esoteric tradition created practices integrating mantra recitation and siddham visualization.

Few scholars have been attracted by the fact that siddham characters, interpreted as iconic signs, were used as the "body" of the absolute language (and therefore) of the absolute reality of esoteric Buddhism.

The Chinese graphic system, and in particular its most ancient characters, was considered as not just a transcription of oral language, but as a system for the representation of reality, constituted by expressive forms in which sounds and scripts are in perfect harmony. In this way, siddham characters acquired the status of microcosms, of absolute entities, in accordance with their myths of origin. Esoteric sources, in fact, consider siddham as absolute, unconditioned, non-created entities. Other sources describe them as spontaneous forms which originally manifested themselves in the sky. In such an absolutist conception of writing, the influence of Taoist elements are also discernible, for instance, the "heavenly talismans" (tianfu) and the "cloud-seals" (yunzhuan). Perhaps, in relation to such myths on the origin of mantric language, Frits Staal suggests that mantras may be sort of fossil vestiges of the process which led to the formation of ordinary language, fragments of the most ancient protolinguistic expressions (1989), "remnants of something that preceded language" (1985:550).


Alper, Harvey P., (ed.), Mantra, Albany: State University of New York Press, 1989.

Bharati, A., The Tantric Tradition, London: 1965.

Comolli, Giampiero, "Nel regno di Amithaba. Su Oriente e simbolo in Jung", in Idem, Rizonanze. Saggi sulla scrittura, il mito, I'Oriente. Rome and Neaples:Theoria, 1993.

van Gulik, Robert H. Siddham: An Essay on the History of Sanskrit Studies in China and Japan. (Savasvati Vihara Series Vol.36). Nagpur: Int. Academy of Indian Culture, 1956 (reprinted 1980, Sata-pitaka Series Vol.247. New Delhi: Mrs. Sharada Rani).

Lopez, Donald S.,Jr., "Inscribing the Bodhisattva's Speech: On the Heart Sutra's Mantra", History of Religions Vol.29, No.4 (1990): p.359

Padoux, André, Vac. The Concept of the Word in Selected Hindu Tantras. Albany: State University of New York Pressl, 1990.

Rambelli, Fabio, "Il gioco linguistico esoterico. Per una teoria del linguaggio del buddhismo giapponese "shingon", Versus. Quaderni di studi semiotici, No. 54, 1989: 69-96

Staal, Frits. "The meaninglessness of Ritual". Numen, Vol.26, No.1: 2-22. 1979a

---. "Oriental Ideas on the Origin of Language". Journal of the American Oriental Society Vol .99: 1 -14. 1979b

---. "Mantras and Bird Songs". Journal of the American Oriental Society Vol. 105 No.3: 549-558. 1985

---. "The Sound of Religion (2)". Numen Vol. 33, No.2: 185-224. 1986

---. Rules Without Meaing. Essays on Ritual, Mantras and the Science of Man. New York: Peter Lang. 1988.

---. "Vedic Mantras". In Harvey P. Alper, ed. 1989: 48-95

Tambiah, Stanley J. "The Magical Power of Words", 1968 (rep. in S. Tambiah, Culture, Thought, and Social Action: An Anthropological Perspective, Cambridge, Mass.: Harvard University Press, 1985: 17-59)

---. Buddhism and the Spirit Cults in NorthEast Thailand. Cambridge:Cambridge University Press, 1970

Toganoo Shoun. Mandara no kenkyu. 1927 (reprinted as Toganoo Zenshu, Vol.4. Kyoto: Rinsen shoten. 1982. Especially pp. 429-469)

Dr. Fabio Rambelli is a researcher at the Institute of Japanese Studies of the University of Venice (Italy). He is the author of numerous publications on the semiotics of Buddhism in Japanese culture.

Go to Semiotic Review of Books Home Page
Go to SRB Highlights
Go to SRB Archives