Introduction: Touching Events in Shared Festivals

1

This essay explores sensory engagement in rituals and its implications for interreligious contact based on participatory observation of shared festivals in Kerala (South India). Kerala festivals embed ritual complexes shared by practitioners of religions other than the one celebrated at the festival venue. This phenomenon of sharing sacred spaces and festivals is, of course, not uniquely South Indian. Sacred spaces celebrated by multireligious communities are common elsewhere in Asia, in the Muslim world, and beyond, often surviving national conflicts and geo-political changes (Gottschalk 2012; Cormack 2012). Such festivals often (if not always) involve sensory religiosity that readily translates from one socioreligious context into another. Ostensibly, this powerful tool in interreligious communication relies on religious polysemy of sensory religiosity.

2

Take, for example, ritual-touching during diverse religious occasions with meanings ranging from blessing (Fig. 1),¹ to healing (Fig. 2), to acquiring merit (Fig. 3), to merely signifying passages between ritual procedures (Fig. 4). These meanings are not inherent in the form (ritual touching) but are rather assigned by convention of all festival participants regardless of religious differences. It may be the case that the same ritual-touching is perceived differently by each religious group.²

3

Postulating ritual as a system of (inter)religious communication calls for developing a framework for description and analysis of ritual typologies beyond the shared festivals in Kerala, with its remarkably lush multireligious landscape.³

4

Sacred spaces in Kerala are normally restricted to members of their respective religious communities; most temples do not allow non-Hindus, some mosques restrict non-Muslims from entry (e. g. Mamburam Mosque), and the Paradeśi synagogue in Cochin is limited to Jews whenever there are enough Jews to perform religious holidays. Even churches that do permit non-Christians to enter are usually frequented by Christians alone.⁴ However, during the dry season, from January-February to April-May, temple, mosque, and church festivals are celebrated at each locality with their own ritual complex variety. These festivals are shared inclusively regardless of religious affiliation, acting as bridges across religious communities.

5

The participants are not merely being entertained in the ritual complex; by way of representing various religious or caste affiliations, they are allotted specific ritual roles.⁵ In such roles, Hindus, for example, are in charge of providing oil for the elephant lamp in St. Mary’s Forane Church in Kanjoor (Fig. 5); or applying oil to the ceremonial cannons at the onset of the festival of the mosque of Kondotty (Fig. 6); or paying homage to the Muslim hunter saint Vāvar as part of the ritual complex of pilgrimage to the Ayyappan temple at Sabarimala (Fig. 7).

6

It is in the context of interreligious contact that sensory religiosity is translated, so to speak, across mythologies and theologies, with otherwise identical ritual procedures acquiring polysemic meanings. I focus on ritual touching for the present study somewhat arbitrarily, aiming at a structural analysis to account for other types of sensory engagement in rituals. Instances of ritual touching facilitate socioreligious communication during festivals that, like pilgrimages and processions, constitute public rituals (Michaels 2016, 211–27). Participant communities, regardless of their verbal intrareligious discourse, communicate their respective religiosities relying on the ritual complex of the shared festival as a platform for non-verbal interreligious discourse. Ostensibly, the shared festivals form a mechanism that is instrumental in balancing religious diversity with cultural cohesion. Kerala has done exceptionally well in this respect.

Religious Diversity in Kerala History

7

Since at least the ninth century, Kerala has been a central juncture of maritime trade between West and East Asia. Its location on the southeastern shores of the Arabian Sea attracted a steady influx of traders and seafarers since times of Antiquity (Gurukkal 2016). Trade was an important source of income for land owners and ruling elites, who encouraged Arabs and Persians to establish market towns adjacent to ports. From the twelfth and thirteenth centuries onwards, diaspora communities of itinerant traders evolved into indigenous monotheist castes (Narayanan 1972, 2013, 277–84; Wink 1997, 275–80; Gamliel 2019).

8

By the time transregional contacts with Arab and Persian traders were forming, Kerala emerged as a regional entity distinguished from Tamiḻakam, the historical ‘macro region’ entity of Tamil Nadu (Veluthat 2009, 295–311; Gurukkal 2010, 242–54). As the sociocultural identity of the Malayalam-speaking region was crystallizing, foreign traders surfaced in the history of the Malabar Coast (Narayanan 2013, 94; see Gurukkal 2010, 261). It seems to me that the regional identity of Kerala owes its emergence, at least to some extent, to these transregional encounters. After all, the earliest document attesting transregional traders in Kerala, namely the Kollam Copper Plates (849 CE), is also the oldest document attesting an early stage of administrative interaction in Old Malayalam (Sekhar 1951, 5; Narayanan 1972, 32–37, 2013, 313–17).

9

Of course, Kerala is not unique in its religious diversity; multicommunal societies feature prominently elsewhere in India. Diversity in unity is a slogan in modern India, which prides itself in its plurality of languages, traditions, and customs. Similarly, shared festivals might very well be instrumental in maintaining socioreligious coherence in multireligious societies elsewhere in Asia, as suggested above. In India, however, sensory religiosity classified as bhakti underlies coherent ritual communication across religions especially in the context of shared festivals.⁶ The density of localities celebrating shared festivals and the myriad forms of performances exhibiting bhakti religiosity are remarkably prominent in Kerala.

10

By contrast, Axel Michaels (2016, 313) denies the role of bhakti in ritual theory, arguing that “rituals are rule-governed […,] public and suitable for an ostentative demonstration of social coherence. […] This is, for instance, different from love, modern art, or faith, including bhakti where besides orthopraxis, a lot of individual, ‘mental’ (mānasa) authenticity is demanded.” I find his argument less than convincing when measured against the ritual practices associated with bhakti religiosity (Ramanujan 1981, 103–69; see also Cort 2002, especially 85–6). Moreover, if sensory religiosity indeed underlies ritual communication, as I hope to demonstrate convincingly below, bhakti ideology cannot be disassociated from ritual and confined to the sphere of individual religiosity, as Michaels suggests.

11

That said, bhakti religiosity, in its social and political respects, regulates and maintains the boundaries between communities and castes in South India in general and in Kerala in particular. Though bhakti conforms to the social norms customary within the Hindu caste system (Burchett 2009), its ideology involves a radical rejection of caste hierarchy and social order, perceived as mundane differentiations obstructing the encounter with the divine. According to A. K. Ramanujan (1999, 327), bhakti “is necessarily an illegitimate relationship, illegitimate from the point of view of law and social order; it is an act of violation against ordinary expected loyalties, a breakdown of the predictable and secure”. It is therefore less than surprising to find bhakti-oriented ideologies and practices across diverse religious communities, as demonstrated in several studies concerning bhakti among Jains, Sikhs, ‘untouchables,’ Christians, and Muslims (Burchett 2009, 2012, 230–46; Doyle 2009; Behl 2007; Cort 2002; Dobe 2010). It remains to be explained, however, how bhakti practices channel interreligious communication and how sensory religiosity is utilized across religious ideologies otherwise exclusive of each other.

12

I therefore approach ritual as a system of communication with underlying sensory semiotics. This approach requires a linguistic analytical framework that is attempted in the following sections. The following section surveys ritual theories based on the analogy between language and ritual, relying mainly on Axel Michaels (2016) and Naphtali Meshel (2014). In “Units of Ritual and Universal Categories in Linguistic Terms”, I turn to Martina Wiltschko’s linguistic universal spine hypothesis (2014) for identifying universal categories of ritual events, while drawing upon E. Thomas Lawson and Robert N. McCauley’s ritual action representation theory (2002b, 2002a). The section “Ritual Events as a Category” categorizes ‘ritual events’ by drawing on linguistic approaches to event structures especially in regard of the semantic-syntax interface between lexical properties of events and their functions in phrasal (or ritual-procedural) combinations. The section “Ritual Competence in Shared Festivals in Kerala” deals with ritual competence as realized in shared festivals, and “Touching Events: Description and Analysis” examines touching events in shared festivals of Hindus, Christians, Muslims, and Jews from a comparative perspective. In the sections “Touching Events and Religions in Contact” and “Sensory Ritual Events and Socioreligious Communication,” I examine emic categories of ritual events in shared festivals in Kerala in relation to ritual events analysis. The latter concludes with a rough sketch of ritual categories, ritual-specific categories, and the prospect of a universal spine of ritual categories. The final section summarizes the discussion with preliminary conclusions suggestive of directions for further research on ritual as a system of non-verbal communication.

Ritual as Language

13

The analogy between language and ritual is based on the assumption that generative rules and deep structures apply for both. This analogy can be traced back to ancient Indian grammarians and ritual theories (Hastings 2003; Meshel 2014, 1–5; Michaels 2010, 2016, 74–77). Grammar in ancient India has evolved out of the need to theorize and systematize Vedic ritual (Kiparsky 1995, 59–60), which may be the reason that South Asianists like Fritz Staal (1979), Richard K. Payne (2004, 2016a, 2016b), and Axel Michaels (2010, 2016) are spearheading ‘grammatical’ theories of ritual. The ritual theorists E. Thomas Lawson and Robert N. McCauley (2002a), too, resort to the language-ritual analogy in formulating a cognitive approach to ritual studies. Their perspective, however, is inclined towards generative linguistics (rather than towards Indic ritual theory and linguistics) in comparing ritual competence with linguistic competence (2002a, 4–6). The language-ritual analogy stems, at least in some respects, from understanding ritual as a system of communication (Edmund Leach, quoted in Michaels 2016, 77–80). Communication necessarily involves various modes of signification, be they verbal or not. The question is to what extent, if at all, linguistic signification is analogous to ritual signification.

14

At least two studies profusely examine the language-ritual analogy. The first is Axel Michaels’ monumental work on Hindu ritual, Homo Ritualis (2016), and the second is Naphtali Meshel’s (2014) minute study of the ‘grammar’ of Biblical sacrifice. Michaels (2010, 2016, 74–117) tests the ritual-language analogy against various models of ritual theories based on years of collaborative research in the Heidelberg School of Ritual Dynamics.⁷ Unlike Meshel, Michaels deals with rituals that are both a living tradition and an indigenous intellectual heritage of ritual theories traced back to Vedic rituals of the second millennium BCE (2016, 293–308). Meshel, on the other hand, deals with the Biblical sacrificial tradition that became obsolete in the remote past. Biblical sacrificial texts and their commentaries exclusively deal with inventories of food offerings to one specific deity in the Israelite temple in Jerusalem before its final destruction in 70 AD. By contrast, the Hindu rituals that Michaels deals with are modular ritual structures across places, periods, and religious traditions (2016, 4 and 25–27). Both studies, notwithstanding these significant differences, share a similar approach to ritual theory based on the language-ritual analogy. Both studies are, therefore, germane to approaching ritual as a system of communication even though both are eventually dismissive of the language-ritual analogy.

15

Michaels distances his analytic framework from linguistics. He states right at the outset that “a one-to-one mapping of the morphology, syntax, pragmatics and semantics of rituals in a manner analogous to linguistic analysis is, to some extent, inadequate” (2016, 94, emphasis added). At the same time, he does promote the language-ritual analogy in stating that “description and analysis of these structures and rules [of Hindu ritual] are nothing else than a grammar, the “‘grammar’ of rituals” (2016, 94). Meshel, too, finally ‘sacrifices’ the grammar of rituals, noting that “little evidence has been found that would justify a high resolution analogy between language and ritual” (2014, 206–9). On the other hand, in his introduction he goes even further than Michaels to suggest a language-ritual homology as the “organizing principle” for his book (2014, 2), while noting the distinctive nature of operative categories in both systems (2014, 2–3, 18–19).

16

Meshel uses the terms analogy and homology interchangeably, notwithstanding the substantially different types of similarity denoted by each term. This terminology, it should be noted, is borrowed into religious studies from evolutionary biology, where the term “analogy” points at genetically-unrelated structures resembling each other, whereas “homology” denotes the similarity of structures of common descent (Smith 1990, 47–48n15; Campany 2018, 335). Therefore, a ritual-language analogy postulates accidental resemblance of structural patterns with no theoretical implications beyond, perhaps, borrowing terms from linguistics to ritual studies. A ritual-language homology, on the other hand, has far-reaching implications for understanding both systems as derived from shared origins. In many respects, it is the latter case that justifies speculations on the ‘deep structure’ and ‘universal grammar’ of ritual; the abstract structures of language and ritual might indeed have a shared cognitive substratum.

17

Both Michaels and Meshel follow previous ritual theorists (e.g. Staal 1979; see Penner 1986) in their approaches to the language-ritual analogy. Neither of them claims that language and ritual structures are of common descent, understandably so. The hypothesis of ritual-language homology is unlikely to be proven or refuted, for tracing an ur-language (or an ur-ritual) would risk involving far-fetched speculations. Yet, the communicative function of and the competence required for both languages and rituals point at a common socio-cognitive ‘deep structure,’ which can be supported by identifying commonalities of structural patterns.⁸ Ritual studies based on the ritual-language analogy provide a good starting point for this type of investigation.⁹

18

Michaels (2016, 80–87) surveys approaches to ritual “building blocks,” seeking to differentiate between rites that are structures and rites that are sections or sub-sections that he terms “ritemes,” “ritual elements,” “actions,” and, following Meshel, also “atomacts” (2016, 82–83). Meshel, too, systematically classifies units, though he further distinguishes between “grammatical” and “ungrammatical” combinations (2014, 132), applying to his analysis tests of syntactic rules commonly used in transformational linguistics. However, Meshel’s inventory of forms relies solely on limited diachronic data,¹⁰ whereas transformational-generative grammar is based on synchronic data drawn from an ever-expanding inventory of forms and structures (see Michaels 2016, 80n13). Michaels, on the other hand, does consult both diachronic and synchronic data, but merely loosely refers to generative-transformational grammar in applying the terms “deep structure” and “universal grammar” to rituals (2010, 10, 2016, 80). Arguably, there is much more a ritual ‘grammarian’ can derive from theoretical linguistics.¹¹ Above all, it seems to me, it is the identification of universal categories across ritual structures that is crucial in the quest after an abstract, ‘deep’ structure of rituals.

Units of Ritual and Universal Categories in Linguistic Terms

19

Admittedly, an exhaustive classification of universal categories of ritual lies beyond my present research scope. However, I suggest two analytical tools as a way forward to identifying the ‘deep structure’ of rituals. The first is the distinction between lexical and functional structural units, and the second is the postulation of ritual event structure. In this, I aim to step back from ritual typologies and zoom in on the micro-level of ritual structures, similar to the micro-analysis of sentence structures in generative linguistics.

20

While descriptive grammars of natural languages provide abundant data for language typologies,¹² they are insufficient in and of themselves for identifying or, alternately, ruling out language universals. The problem, according to Martina Wiltschko (2014, 23), is that typologies yield evidence for both the universality of categories and for variation in the inventories of categories in diverse languages. To resolve the tension between universality and variation, Wiltschko (2013, 2014, 23–29) postulates a universal spine of categories divided between verbal and nominal domains. If we apply the same method of categorical classification to ritual typologies such as those in Michaels (2016), Payne (2004, 2016a, 2016b), and Meshel (2014), we may identify universal as opposed to ritual-specific categories. Importantly, as Wiltschko argues for language categories, the variation in ritual-specific categories does not, in and of itself, rule out the possibility of identifying abstract, generic patterns underlying variation and diversity.¹³

21

Ostensibly, the category of ‘ritual actions’ is the most likely candidate for constituting a universal category. I would, however, rather use the term ‘ritual events,’ for reasons explained in the following section. This category is evident in Michaels’ survey of several ritual typologies highlighting ritual event as a recurring category across the board from ritual action in Lawson and McCauley (2002b) to atomact in Meshel (2014, 130). For Michaels, this is perhaps the most evident category, as he has at his disposal a rich inventory of emic categories derived from verbal nouns (e.g. saṃkalpana or saṃkalpa, resolution, sambodhana, addressing), or from nouns implying actions (e.g. pradakṣiṇā, circumambulation, and dakṣiṇā, sacrificial payment; Michaels 2016, 43, 96–97; see Payne 2004, 2016a, 2016b).

22

While many ritual traditions develop a metalanguage to refer ‘grammatically’ to ritual categories, they do not necessarily categorize ritual events in a similar manner and to the same extent that the Indic ritual tradition does. However, event categories can be postulated for any given ritual tradition, just as grammatical categories can be postulated for any given language, regardless of whether the speech community has its own indigenous grammatical tradition or not. Wiltschko’s two propositions underlying her universal spine hypothesis are enlightening in this regard. She states as follows (2014, 24):

23

1. Language-specific categories (c) are constructed from a small set of universal categories K and language-specific UoLs [Units of Language]
2. The set of universal categories K is hierarchically organized where each layer of K is defined by a unique function.

24

The first proposition (i) can account for ritual categories; it can be assumed, at least tentatively, that ritual-specific categories are constructed from a set of universal categories underlying ritual-specific elements, such as those listed and described in ritual typologies. The second proposition (ii), too, can be applied accordingly to ritual theory, though we still need to identify the set of universal categories of ritual to work with.¹⁴ To develop a tentative set of universal ritual categories, I take cues from both linguists and ritual theorists and suggest the category of ‘ritual event’ as a tentative universal category of ritual.

Ritual Events as a Category

25

Thomas E. Lawson and Robert N. McCauley (2002b, 2002a) identify the categories “action,” “agent,” and “patient” in their theory of action representation system underlying the cognitive and psychological bases of religious rituals.¹⁵ They state that actions constituting ritual structures are familiar, even trivial, actions in the sense that they require the perception of agents and patients. Ritual actions, however, while involving agents and patients, are “inevitably connected sooner or later with actions in which CPS [culturally postulated supernatural] agents play a role” (2002a, 159). While the role of CPS agents may be challenged (Keane 2008, 115), the analytical model of the action representation system is germane to approaching ritual structures for their ‘grammar’ and, in particular, for identifying ritual event categories. Lawson and McCauley maintain that “the representation of religious rituals requires no special cognitive apparatus beyond the garden-variety cognitive machinery all normal human beings possess for the representation of agents and their actions” (2002a, 11). Basing ritual cognition on a par with human cognition of agents and actions is suggestive of a cognitive core underlying both ritual and language categories of agents and actions. In other words, Lawson and McCauley’s theory supports the hypothesis of language-ritual homology.

26

Lawson and McCauley’s analysis seeks to “capture familiar presumptions about the internal structures and external relations of actions too […]. [W]hile cognitive scientists have proposed interesting accounts of our understanding of agency, they have had much less to say about our understanding of actions” (2002a, 11–12). Nevertheless, linguists do propose detailed and rich analyses of actions, or, more accurately, ‘events’. In linguistics, the notion of an action requiring an agent is but one sub-category of ‘events.’ The category of ‘event’ embraces a broader range of categories besides agents, patients, and objects, extending to modifiers such as temporal and spatial adverbs. Events in linguistics are further analyzed for their lexical properties (e.g. volition, motion, stativity, causativity, passivity), argument structure (e.g. agent, theme, patient, instrument), and temporal and spatial properties (e.g. tense, aspect, deixis, prepositional phrases). Event categories can even be grammaticalized to modify other events and to function as light verbs or auxiliary verbs.

27

Arguably, ritual events contain these four characteristics of lexical properties, argument structures, temporal/spatial properties, and auxiliary functions (acting upon other ritual events). It is this characterization of ritual events that the present discussion relies on in support of the language-ritual analogy; the phenomenon of event-modifying verbs is of particular interest here as it hinges on a universal pattern of multi-functionality, which is, as an anonymous reviewer noted, a diagnostic tool for relating language-specific categories to a universal structure. Event-modifying verbs, namely auxiliary verbs, exemplify multi-functionality according to Martina Wiltschko (2014, 3).¹⁶

28

At this point, a few words on the distinction between lexical signification and syntactic function are in order. Consider a simple referential statement like “the cat sits on the mat.” The words ‘cat,’, ‘sit,’ and ‘mat’ signify; they have lexical properties that denote perceptible objects (nouns) and occurrence (verb). But these three units—two nouns and a verb—cannot refer to any actual situation in the real world without the non-lexical segments—the articles, preposition, and tense marker; these sentence segments relate the three lexical units to a specific situation denoted by the complete utterance. Without these function words, the sequence of the lexical words ‘cat,’, ‘sit,’ and ‘mat’ would fail to communicate a situational meaning. This basic distinction between lexicon and grammar and, consequently, between lexical and functional units is essential for identifying categories and sub-categories of linguistic units.

29

The Piercean categories of symbols and indexes are useful in applying the lexical-functional distinction to ritual theory. Clearly, ritual signification lacks audible phonetic substance (words and morphemes) but rather operates within non-verbal semiotic systems.¹⁷ In other words, ritual events can be conveniently divided between symbols and indexes in parallel to the division between lexical content words and grammatical function words. A ritual event is functional when it modifies another ritual event, similar to auxiliary verbs such as ‘have’ and ‘be’ that modify the event structure of main verbs. Ritual communication, too, relies on various combinations of these two types of symbol and index signification. As demonstrated below (“Touching Events: Demonstration and Analysis”), functional ritual events (indexes) derive from content events (symbols) by shedding off certain lexical properties of the latter, namely, by way of semantic ‘bleaching.’ This distinction between lexical and functional ritual events facilitates the structuring of rituals, just like content and function words structure utterances. It parallels the pattern of contrast (Wiltschko 2014, 6–10) alongside the pattern of multifunctionality, calling for further engagement with the ritual-language analogy.

30

Lawson and McCauley address the lexical-semantic unit of ritual (action) but leave out the functional-syntactic one: “many actions [i.e. ritual events] that religious persons repeat in religious ceremonies (such as everyone standing at certain points in a religious service) will not count as rituals” (2002a, 14). While indeed not all events that occur during rituals belong to the ritual structure, it seems to me that “standing at certain points in a religious service” signifies precisely what I would categorize as a ritual event, a ritual standing-event. It remains to be asked, however, what precisely its internal structure is and how it relates to the ritual structure as a whole (as index or symbol). To my understanding, this distinction is necessary for analyzing ritual as a socio-cognitive system of communication.

Ritual Competence in Shared Festivals in Kerala

31

Lawson and McCauley (2002a, 4–6) convincingly argue for incorporating the notion of competence in ritual theory, yet again resorting to a concept developed in linguistics, “because of the striking similarities we noted between speaker-listeners’ knowledge of their languages and participants’ knowledge of their religious ritual systems.” Ritual competence in shared festivals, though, is not confined to the religious ritual system of a certain group; it is rather manifested by participants of different religious affiliations. It thus facilitates the mechanism balancing socioreligious diversity with communal segregation. Presumably, ritual competence can also facilitate other, less benign forms of interreligious (mis)communication, which need not concern us here.

32

The shared festivals examined in the present study demonstrate a non-verbal discursive competence between otherwise segregated religious communities. If ritual events are the conveyors of meanings, whether lexical or functional, then it is their differentiation from ‘real world’ events that enables this multireligious sort of ritual communication.¹⁸ Presumably, the sensory semiotics typical of bhakti religiosity underlies the ritual competence that allows for this sort of communication. However, as I argue below, the semiotics of sensory religiosity is often, perhaps even mostly, derived from a generalized religious engagement far from confined to South Asian religiosity. Ritual competence is an intuitive capacity to differentiate between a ‘real life’ sensory event and its ritual parallel, for example eating an apple as opposed to eating a consecrated substance, even if the observer is clueless about the religion of the participants in the ritual event. An analysis of sensory semiotics in rituals should, therefore, be based on an analysis of the internal structure of ritual events.

33

The internal structure of ritual events is described and analyzed in more detail below, focusing on ritual events of touching.¹⁹ The analysis of ritual touching events is based on the context, content, and function as well as on the lexical properties and argument structure of each touching event. Presumably, sensory semiotics is constantly employed in ritual signification across religions,²⁰ rendering ritual competence analyzable regardless of religious and cultural boundaries (Lawson and McCauley 2002b, 176, 2002a, 10; Barrett 2000, 33).²¹ Therefore, ritual events based on sensory semiotics are the best-fitting for a preliminary examination of their lexical as opposed to functional properties.

Touching Events: Description and Analysis

34

The touching events demonstrated above (Fig. 1–4) are revisited below for scrutinizing their lexical versus functional properties. The touching event presented in Fig. 4 is, as stated above, commonly observed by devotees upon exiting the temple.

35

Every devotee touches substances—water, powders, ashes, and flowers—consecrated earlier to the temple deities and presented at the spot near the temple doorway just before leaving the temple grounds. This touching event is a functional unit marking the completion of the temple worship. Therefore, I term it ‘completive,’ compared with the ‘accomplishment’ aspect for its temporal properties of telicity, duration, and dynamism (see Smith 1999, 481). In terms of its argument structure, this ‘completive touching’ event has the devotees themselves get in touch with the ritual substances, with no direct involvement of other entities such as the priest or deity. Moreover, the touching is disassociated with attributions such as healing, blessing, etc. that are typical of content units. In other words, the functional-syntactic units of ritual undergo semantic bleaching.

36

Touching events of the lexical-semantic type, unlike the functional-syntactic, contain semantic content by way of attribution of ritual efficacy such as blessing, healing, merit acquisition, and so forth. The attribution of ritual efficacy in these cases is provided by an etiology linking each touching event with its respective ritual efficacy. Thus, in Fig. 1, the Teyyam oracle features a deity called Toṇṭaccan (Namboothiri 2005, 15–16). As a rule, Teyyam oracles are apotheosized low-caste persons who died a violent death at the hands of oppressive upper-caste people. As a result, they become fierce and dangerous deities, threatening to consume the health and prosperity of the oppressor’s clan. To avert the danger, the landlord and his clan perform an annual possession ritual featuring the apotheosis story in song and dance that invokes the deity through the oracle dressed in an elaborate, heavy costume (Menon 1993; Freeman 2005; Vadakkiniyil 2010). In Fig. 1, the oracle has just finished dancing with fire; it is past midnight. The members of the extended family funding the ritual line up to receive personal blessings for a prosperous year.²²

37

As the touching event in Fig. 1 signifies blessing, a content unit of ritual, it can be analyzed for its lexical (rather than indexical) properties. Firstly, it is a volitional and transient touching event; its argument structure has an intentional agent (the oracle) and a recipient or object of the blessing (the devotee). Secondly, the argument structure requires two entities (comparable to noun phrases in linguistic analysis). Thirdly, the aspectual properties (aktionsart) of the blessing-touching event are comparable to the completive touching event; they are dynamic, telic, and durative.

38

It is possible to further analyze this blessing-touching event to derive a higher resolution of the ritual-specific properties involved. Thus, for example, this blessing-touching event has constituent ‘phrases,’ such as an exchange of a currency note from the devotee for a flower plucked out of his belt by the oracle. Such constituents are ritual-specific categories comparable to the language-specific categories that constitute phrases in linguistic structures. However, for the present discussion I prefer focusing on a comparative analysis of the rough contours of touching events under consideration for the sake of an overarching classification of the contrast between indexical and content units.

39

The touching event depicted in Fig. 3 is a lexical-semantic unit in an extensive ritual complex—the ten-day annual festival honoring St. Sebastian in the St. Mary’s Forane Church in Kanjoor (Central Kerala). As the story of St. Sebastian’s violent death is well known, it need not be repeated here. What is less known, perhaps, is the power attributed to golden arrows (associated with the saint’s martyrdom) to transmit merit upon touch. Decorated paper plates with a golden arrow and cross (Fig. 8) are presented on long tables at the entrance to the church. The devotees pick up a plate and circumambulate the stone-cross pillar facing the entrance to the church, while holding the plate in their left hand and touching the arrow with their right hand. According to the local tradition, this is an offering beneficial to any person (regardless of caste and religion); it helps with getting married, begetting children, passing exams, getting a good job, buying land, and curing terminal diseases. In terms of ritual efficacy, the meritorious power of the arrow originates in the Christian theology of martyrdom. But it is polysemic, as the martyrdom of the saint in the context of Kerala is readily translatable into the apotheosis of Teyyam deities, where weapons are transferred to the oracle during the dedicatory rites, serving as props all through the possession ritual complex.

40

In terms of event structure, touching the arrow is a state; it is durative, static, and atelic, for the merit gained is indefinite. It is a complex event subordinating another event of circumambulation. It is volitional and intransitive; the devotees intentionally get in touch with the arrow rather than touching an object arrow.

41

But there is, I believe, an additional and less obvious significance underlying the tangible contact with the arrow. The festival features two ritual events with the arrow as a ‘constituent’ phrase; one is the arrow-offering circumambulation (ambә eḻunnuḷḷikkal) shown in Fig. 3, and the other is the adorning of St. Sebastian’s idol with arrows and crown made of gold in a ritual event called ambum muṭiyum cārttal, ‘arrow-crown ornamenting’ (Fig. 9). This type of contact between consecrated objects and an idol is an empowering one, signifying the onset of immanence, as if waking up the latent presence of the saint from a frozen state of transcendent slumber for actively moving across the town in processions during the major festival days (January 19 and 20).

42

Empowering by getting divine entities in contact with consecrated weapons (normally, swords and daggers) is common in possession rituals in Kerala.²³ Weapons are thus incorporated into sensory religiosity. In the Kondotty mosque festival, a ritual event called tōkk-eṭukkal, ‘canon firing,’ is performed at an open ground near the mosque, signifying the beginning of the festival processions (peṭṭi-varavә). Before the canons are carried to the open ground, they are placed near the entrance gate of the takkiyāvә,²⁴ where devotees gather around to smear the cannons with oil donated by groups of various religious affiliations (Fig. 6). This is a touching event signifying the inception of the cannon-firing ceremony. The devotees touch the object of veneration (the canons) using consecrated oil for the instrument. It is an inceptive touching event with a ditransitive argument structure, and dynamic, durative, and atelic temporal properties.

43

The cannons have historical, religious, and social significance; they were abandoned at the battlefield by the army of the Zamorin of Calicut, after his defeat in the battle of 1791 (Dale and Menon 1978, 531; Sattār 2009, 28). The cannons are attributed with ritual efficacy, as people attribute healing power to the cannon-oil, believed to cure skin problems. The social significance of the cannons is manifested through the allotment of ritual roles to Muslims and Hindu Dalits in polishing, carrying, and firing the cannons (Thangal 2009, 35–36; Sattār 2009, 28). In terms of universal categories as opposed to ritual-specific categories, weapons as an argument in ritual-event structures can be construed as a ritual-specific category in shared festivals in Kerala.

44

Similarly, applying oil to a consecrated object and collecting oil from the same object is a common constituent in touching events. During the St. Sebastian Feast, oil enacts a historical bond between the local Hindu and Christian communities in causative relationship with enhanced valency. Hindu participants continuously pour oil over the elephant lamp (Fig. 5), and devotees continuously gather it from the lamp, after it passes through the bronze elephant and the two figures riding on it.²⁵ The Hindus pouring oil thus act as agents in a causative healing touch; they cause the oil poured over the lamp (instrument-argument in syntactic terms) to get in touch with the devotees. The elephant lamp thus signifies the historical Hindu royal patronage over the Christian church.

45

Like the cannons in Kodotty, the elephant lamp in Kanjoor is associated with a local historical account. According to one version of the story (there are quite a few), King Śaktaṉ Tamburāṉ (Rama Varmman IX, 1751–1805) once visited the Bhagavati temple Putiyeṭam Kovilakam, located within earshot from the church.²⁶ At the time, the Feast of St. Sebastian was celebrated. The king, disturbed by the sound of exploding fireworks, ordered his men to confiscate the church’s fireworks and throw them into the river. As the saint is so powerful, the river could not contain the fireworks, which surfaced afloat. Soon after, a terrible headache seized the king. Realizing that the saint was offended, the king’s Brahmin minister advised him to appease him. So the king offered the elephant lamp (Fig. 10) to the church and was cured of his headache. Ever since then, the oil poured into the lamp is attributed with the ritual efficacy of healing, specifically bodily pain. Touching the oil poured over the elephant lamp is, therefore, a lexical-semantic type of ritual touching events. Unlike the inceptive touching event in Kondotty, the healing touch of the elephant lamp in Kanjoor does not signify any other event during the festival; people continue pouring the oil, touching it, and collecting it all through the festival days. It is thus a durative, atelic, and dynamic event with enhanced valency.

Touching Events and Religions in Contact

46

The touching events discussed above demonstrate how sensory events in rituals readily adjust to varying religious contexts and channel interreligious communication. The contact with certain objects, like the Kanjoor elephant lamp and Kondotty cannons, triggers historical interreligious bonds. But some ritual events also enact these historical interreligious bonds. An example for a touching event that enacts a Hindu-Muslim bond is the ritual event called candanakkuṭa, ‘sandal-paste pot’ or candanam-eṭukkal, ‘carrying the sandal paste.’²⁷ In Changanssery (South Kerala), the procession with the sandal-paste pot proceeds from the mosque towards the neighboring church and temple, honored by dignitaries of all the neighboring communities as it passes by (Haneefa 1999). Rahman Thangal (2009, 35–36) describes the ‘carrying the sandal paste’ procession (Fig. 11) as follows (the superscript letters represent word origins from Arabic^A, Persian^P, Turkish^T, Sanskrit^S, and Malayalam^M):

47

The rite (caṭaṅṅә^M) called “candanam^S-eṭukkal^M” is the most important one. It is perfused with bhakti^S and highly nuanced. The sandal paste of the utmost quality, especially ground on the foundation stone of the mosque, and a food offering (prasādam^S) called marīda^P?,²⁸ which will later be distributed in the mosque, are placed into special vessels. The chief Thangal (taṅṅaḷ^M)²⁹ and his representative carry [these vessels] on their heads and bring them over to the mosque (khuba^A,قبة ), walking barefoot. This is the ritual (karmmam^S) called candanam-eṭukkal. They set forth from the chief Thangal’s thatched shed, uttering a special prayer in the takkiāvә^T. They first take the sandal paste and the marīda from the old storeyed house of the Thangal’s family to the base of the flagstaff and place it there for some time. They chant some Qurʾān verses and sprinkle rose water over the disciples standing in front of them. Thereafter, the chief Thangal and his representative, carrying the sandal paste and the marīda on their heads, proceed to the mosque in an ambience saturated with bhakti, along with a retinue of disciples and soldiers, who chant Arabic poetry (bayyittә^A بيت), maulid^A songs and praise (madh^A مدح gānam^S) songs.³⁰ At that time, the Thangal and his representative must not put on shoes. They are being venerated with royal parasols. At the entry gate of the mosque, specifically ordained heads of families (kāraṇavanmār^M) receive the sandal paste and the marīda and take it into the mosque. It is only after washing their feet (vuḷәhә^A وضوء  eṭuttә^M), that the Thangal and his representative enter the mosque. While the chief Thangal and his disciples go forward for lifting the sandal and the marīda and enter the mosque, the big drum called nakāra^P (نقاره) is beaten. During that time, the soldiers stand put at the entry gate of the mosque. It is in the tomb (khabaṟә^A قبر) of Hazrat^P/A Muhammad Shah^P [the founder of the mosque] inside the densely crowded mosque that the immediately following rituals take place.

48

What is striking about Rahman Thangal’s description is the rich and complex structure of the touching event candanam-eṭukkal, involving diverse entities and “arguments” (agents, patients, objects, places, times) as well as sub-events. Contact with the consecrated sandal paste ignites, so to speak, a complex web of interreligious and cross-cultural contacts, as the multilingual nomenclature combining Arabic, Persian, Sanskrit, and Malayalam words suggest.

49

The canadanam-eṭukkal of Kondotty is in fact a relatively small and quiet procession, especially in comparison with the procession immediately preceding it, the taṭṭāṉ-peṭṭi varavә, ‘the jeweler-box procession’ (Figs. 11 and 12), which is the grandest procession of all the processions (varavә) celebrated throughout the week of the festival. Similarly to the role of the Hindus pouring oil into the elephant lamp in the St. Sebastian festival in Kanjoor, in Kondotty, too, Hindus are assigned an active role during the mosque festival. The jewelers’ (taṭṭāṉ) participation in the Muslim festival is attributed to a time when the Jewelers were hit by a plague and cured by the first Thangal’s healing powers. The jewelers’ historical gratitude is marked by an object of silversmith craftsmanship carried through the procession—silver banners (Fig. 12) that are displayed near the tomb of the first Thangal of the mosque, Hazrat Muhammad Shah.³¹ This procession includes typical Hindu festival performance arts, such as procession dancers and drummers (Fig. 13), as well as several types of touching events, such as self-immolation touch (Fig. 14) and healing touch (Fig. 15), possibly reflecting well-known Shiʿa practices.

50

Before concluding this section on touching events, an example of a touching event in the Kerala Jewish festival is in order, for what makes Kerala’s diversity remarkably unique is its sustenance of a small Jewish minority for a few centuries prior to the colonial period (Gamliel 2018b, 2018a; see Wink 1987; Segal 1993; Katz and Goldberg 1993). Though the Malayali Jewish community is almost completely gone, the celebration sealing the high holidays, Simḥat Torah, ‘Feast of the Torah,’ was miraculously kept alive even as late as 2016 (Rajoo 2016). As the only synagogue in Kerala to maintain the celebration is the Paradeśi synagogue in Mattancherry, my description and analysis of the touching event and its significance in interreligious communication rely on the celebrations I witnessed there in 2004 and 2007.

51

The Jewish Holidays are not a shared festival in and of themselves. On the contrary, non-Jews are forbidden from entering the synagogue during prayers and holidays. In the past, there used to be a Jewish festival in a sacred place that is still remembered among Kerala Jews in Israel by the name nērcca, which is the same term used by Muslims in Kerala to refer to their mosque festivals. The Jewish nērcca used to be celebrated around the tomb of Nāmya Mutta, a poet-saint of the sixteenth century (Walerstein 1987, 95–113, 2006). This nērcca must have been an occasion of interreligious sharing of the sacred, for the tomb is still preserved and honored (by lighting candles) by the Muslim, Christian, and Hindu neighbors living around it (Fig. 16).

52

Since we have no detailed ethnographic descriptions of the Nāmya Mutta nērcca, it is impossible to identify the ‘building blocks’ of the ritual structure of this festival, let alone describe and analyze touching events therein. However, hints regarding the interreligious significance of touching events in Jewish festivals in Kerala can be found in Simḥat Torah, as it ends in a celebration shared with the non-Jewish neighbors of the community. Despite the fact that the Paradeśi synagogue remains closed to outsiders all through the High Holidays, once the last round of circumambulations with the Torah scrolls ends, the women enter the main prayer hall (normally reserved for men alone), and the gates of the synagogue’s compound are opened, inviting non-Jewish neighbors to enter and watch from the windows and women’s gallery as the completive ritual of entering the Torah scrolls back into the ark takes place. It is this final ritual that involves a touching event; the women of the community line up before the ark to touch each scroll with hands and lips (Fig. 17). After this ritual touching, the men close the ark and swiftly start dismantling the maṇāṟa (literally, wedding chamber), decorated with jasmine flower garlands, silk garments, and shining lights.³²

53

This ritual touching is a completive event; it signifies the completion of the Jewish holiday, marking a passage to a shared celebration which the non-Jewish neighbors are invited to join. A tree-shaped lamp (āl-viḷakkә) is lighted outside the main gate of the synagogue compound, and the Jews go out in a jolly procession accompanied by their non-Jewish neighbors. They all walk to the house of the eldest male member of the community, the mutaliyār, for food and drinks until the late-night hours. It is perhaps due to this type of ritual sharing that the non-Jewish neighbors of the last Jews of Cochin still strive to maintain the Simḥat Torah celebration.

54

In 2008 and 2009, I took part in the Simḥat Torah celebrations in Mesilat Zion, one of the villages of Kerala Jews and their descendants in Israel. The same customary concluding rituals are still practiced there, though there is no occasion for interreligious contacts in the predominantly Jewish state of Israel. Nevertheless, the feast concluding the High Holidays is still celebrated, and its categorization differs slightly from the one used in the Paradeśi context. The Malayalam-speaking Jews in Israel refer to the eldest male member by the term onnāṉ-kārṇōrә, ‘foremost male elder,’ rather than mutaliyār, ‘leader.’ This is significant because the term kārṇōr (<kāraṇavar) is typical of the matrilineal kinship system in premodern Kerala, predating the term mutaliyār. The latter title was bestowed by the Dutch East India Company (VOC) on the leader of the Paradeśi community in the late seventeenth century. Thus, the onnāṉ-kārṇōrә feast bears testimony to a past of shared festivities shaped by interreligious contacts. Remarkably, even in Israel the completive touch is still performed just before the congregation exits the synagogue to celebrate the onnāṉ-kārṇōrә feast (Fig. 18).

Sensory Ritual Events and Socioreligious Communication

55

The description and analysis of touching events in the previous section is aimed at investigating sensory religiosity in ritual. But not all ritual events are based on sensory engagement, even though most of them, at least in the context of shared festivals in Kerala, engage the senses to varying degrees. Moreover, touching events is an analytic category that only partially overlaps with emic categories, and even then, the emic categories are classified differently. That said, the category ‘ritual event’ subsumes emic categories, such as those attested in leaflets for or oral descriptions of shared festivals. The organizing committees of festivals often produce leaflets with a timeline of events for each festival, listing event categories like ‘flag-raising’ and ‘canon-shooting.’’ These are ritual-specific events with terms conventionally used within the community.³³ The leaflet in Fig. 19, distributed among participants of the Kanjoor Church festival, is an example for such indigenous typology of ritual events.

56

Upon close examination, the ritual events outlined in the leaflets are, in most cases, complex events; they can be ‘parsed’ into smaller units of event categories and ‘glossed’ accordingly (e.g. blessing, healing, inceptive, completive). Recall that a unit—whether lexical or functional—is an event containing at least one entity.

57

Michaels (2016) and Payne (2004) recognized certain structural patterns of rituals that are evident also in the Malayalam leaflet (Fig. 19), for example repetition, as with the special mass (viśeṣaṇa kurbbāṉa). The special mass is repeated every morning, but the pattern of repetition is not a structural unit; the special mass is a complex ritual event that needs to be further analyzed for its subordinate event categories. Some of these subordinate events are listed as well, like the novena prayer (novēṉa^PO) and the sermon (prasaṃgam^S).³⁴ Other subordinate events are known by convention, like eating rice grains (prasādam^S)³⁵ offered at the entrance and the Eucharist (kurbāṉa^SY) offered during the mass.

58

Leaving aside the lexical meanings of each subordinate event, the sensory quality of many of the ritual events becomes obvious. Thus, the first two ritual events (novena and sermon) constitute a hearing event, whereas the prasādam and the Eucharist constitute a tasting event. The special mass concludes with the sacristan (kārmmikaṉ^S) crossing his chest with his fingers, while chanting eppōḻum, ippōḻum, ennēkkum (any time, even now, forever), lowering the pitch of the last syllable of each word. This concluding gesture is a passage to the specific ritual events scheduled for each of the ten festival days (January 17–27 annually). Notably, the special mass is ‘vernacularized’ by juxtaposing ritual events such as prasaṃgam and prasādam with the pan-Catholic categories of novena and the Eucharist respectively.

59

Some of the festival days (January 17–21 and 26–27) are particularly important; festive processions and rites follow the daily special Masses. One such rite is the koṭikayaṟṟam, ‘flag-raising,’ scheduled for January 17. It marks the beginning of the festival like in many other shared festivals in Kerala.³⁶ It is, therefore, an inceptive event marking the onset of the ritual complex.³⁷ Notably, the leaflet also lists the ritual entities participating in the events, in this case the name of the officiating priest (vikāri^PO) in charge of the litany (ladīññә^PO) that follows the flag-raising event.³⁸ Once more the analysis of the ritual event structure reveals the fusion of Kerala-specific and Christian-specific ritual categories. Seen from this perspective, the shared festival is structured like a hybrid language register, which possibly emerged alongside hybrid linguistic registers like Maṇipravāḷam and Arabic-Malayalam.

60

Similarly, the Kondotty mosque festival (nērcca) is announced by a flag-raising event (koṭiyēṟṟam). It is an elaborate event branching into several rites and rituals as described by Rahman Thangal K. T. (Thangal 2009, 34–35):

61

The beginning of the rites starting the nērcca^M is flag-raising. This rite would be announced around the harvest season, attracting the disciples (murīd^A) and devotees (bhakta^S) of Thangals from faraway regions to the Thangal of Kondotty. It is only then that the date of the nērcca is fixed. The flag-raising is on the tenth of an Arabic month. Once this is done, the nērcca starts on the thirteenth of the next [Arabic] month. On the afternoon of the flag-raising day, there is a special prayer (prārthana^S) at the mosque (khubba^A). The disciples and regional representatives proceed from the revenue-office building to the mosque in a procession (ghōṣayātra^A), carrying white flags. After praying there, they proceed to the takkiyāvә^T. Then, the senior Thangal and his attendants go to pray in the mosque, return to the takkiyāvә^T and return with the procession retinue to the flagstaff (koṭimaram^M). At the flagstaff base, they first sign a letter of invitation to the nērcca^M. Copies of this letter are sent to business leaders, disciples, and community leaders. The Thangal dips his hands in sandal water and impresses them on two flags. On one flag [he impresses] thirty-five [hand imprints] (for the shaikhs [mashāyikh^A مشائخ] of the order [tvārīkhattә^A طريقة]); on the other [flag, he impresses] eleven [hand imprints] (for the Islām اسلام and īmān^A امان articles of faith). As soon as the Thangal goes to the mosque for a prayer (duʾāʾ^A دعاء) and upon his return, they beat the nagāra^P drum. After that, people especially assigned [for this task] raise the white flags on the flagstaff. On the next day, during daytime, they raise the flag on a small flagstaff behind the mosque. On the days immediately following, they distribute the imprinted flags to be raised in the nearby places Kanjiraparambu, Nambolamkunnu, Melangady, and Kaloth.

62

Compared with the Kanjoor Church flag-raising event, the Kondotty mosque flag-raising is an elaborate complex event. Particularly noteworthy is the touching event of imprinting flags with sandal water as if in anticipation of the sandal-paste pot procession concluding the festival. Despite the high degree of variation between the church and the mosque festivals, both flag-raising events are similar in function (inceptive), with religion-specific categories (ladīññә, duʿāʾ) and sensorial engagement (beating drums, dipping hands in sandal liquid) featuring important religious officials as syntactic arguments embedded in the event structure.

63

To summarize, the constituents of ritual-specific events like the special mass and flag-raising are subordinate events with their own event structures, often involving sensory religiosity. In order to approach the structure of rituals as a non-verbal linguistic structure, the level of the single ritual event structure is the best-fitting for analysis, especially if we wish to derive a sound socio-cognitive analysis of ritual structures.

From Ritual-Specific Categories to Categorical Hierarchies

64

Several preliminary conclusions can be drawn at this stage. Firstly, it is possible to generalize ritual-specific events (e.g. blessing, praying, divination, healing, offering) based on patterns of multi-functionality and contrast (see Wiltschko 2014, 3–10). Secondly, at least one category of sensory events (e.g. touching events) can be construed religiously as ritual-specific events. Thirdly, at least two universal ritual categories, events and entities, can be identified as the ‘building blocks’ of ritual units across religious traditions. Unlike Lawson and McCauley (2002a, 8–11), however, I do not postulate agency as a prerequisite to ritual events but rather assume event structures, some of which may not have an agent in their argument structure. Fourthly, ritual event structures, as suggested above, can be analyzed for their respective argument structures, lexical properties, and temporal-spatial properties. Fifthly, the attribution of ritual efficacy to ‘real-life’ events results in construing the same events—along with their internal structural constituents—as ritual-specific categories.

65

As linguists show, the event structure, in terms of its lexical properties, involves all the elements, including the arguments, modifiers, and temporal properties (Verkuyl 1989; Pustejovsky 1991; Rappaport-Hovav, Doron, and Sichel 2010). Likewise, in ritual event structure analysis, events encompass arguments, modifiers, and temporal properties, as in “walk X times around Y” (e.g. circumambulation, pradakṣiṇa), or “eat Y at time of X” (e.g. the Eucharist, prasādam). Notably, the structure of ritual events is necessarily bound to ritual-specific times and places. All the touching events examined above, for example, are bound to predetermined and fixed spatial arguments within the ritual complex. Every ritual complex, with all its constituent events, is thus bound, differentiating in this way its event ‘system’ from its non-religious event ‘environment’ (see Krech 2018). It is, therefore, evident that identifying spatial and temporal properties is a prerequisite to ritual structure analysis. This differentiation explains to a certain degree the process of attributing ritual efficacy to otherwise ‘real-life’ events.

66

Admittedly, the present discussion can only go as far as to suggest that universal categories of ritual structures exist in the same way they exist for languages. I propose a universal category of ritual events underlying specific-ritual categories identified on the basis of multi-functionality and contrast following Martina Wiltschko’s (2014, 3–10) approach to identifying linguistic categories. Furthermore, I propose to approach rituals as a complex of event units, some of which may subordinate smaller event units, and to zoom in on the smallest possible units of ritual events in order to identify and classify their constituents and internal event structures, as is done for ‘real-life’ events in lingusitics (see Butt and Gueder 2001; Laca 2004; Verkuyl 1989; Pustejovsky 1991; Rappaport-Hovav, Doron, and Sichel 2010). Extensive typologies of ritual structures may thus be tested for universal categories based on syntax theory and generative linguistics. Further research in this vein may reveal a deeper socio-cognitive analogy (perhaps even homology) between language and ritual.

References