Journal of Pentecostal Theology 20 (2011) 122–154

brill.nl/pent

New Frontiers in Tongues Research: A Symposium Randall Holm, Matthew Wolf, and James K.A. Smith* Editorial contact: James K.A. Smith Calvin College, 1845 Knollcrest SE, Grand Rapids, MI 49546, USA [email protected]

Abstract Theological reflection on the ‘lived reality’ of Pentecostal/charismatic spirituality continues to consider tongues-speech ( glossolalia) an important phenomenon that still elicits new questions and research trajectories. In this interdisciplinary symposium, the authors consider the ‘state of the question’ in theological, philosophical, and linguistic research on tongues and offer new proposals and paradigms. Keywords glossolalia, linguistics, naturalism, Abraham Joshua Heschel, Martin Buber

Tongues as a Blush in the Presence of God Randall Holm This paper follows up James Smith’s contribution at the 2006 SPS conference, ‘Tongues and Philosophy of Language: Conceptual Production at the Limits of Speech’,1 where Smith moved the conversation on ecstatic tonguesspeech from asking, ‘What does tongues mean?’ to ‘What does tongues do?’ Continuing with an emphasis on doing over meaning, in this paper I am pressing the ontological button and revisiting the primordial question, ‘What then is tongues-speech?’ In particular I argue that if tongues-speech functions * Randall Holm (PhD, Laval University) is Associate Professor of Biblical Studies at Providence College in Manitoba, Canada. Matthew Wolf (MDiv, Trinity Evangelical Divinity School) is Assistant Editor of Bibles and reference works at Tyndale House Publishers in Carol Stream, IL. James K.A. Smith (PhD, Villanova University) is Professor of Philosophy at Calvin College in Grand Rapids, MI. 1 Subsequently expanded and published as James K.A. Smith, ‘Tongues as Resistant Discourse: A Philosophical Perspective’, in Speaking in Tongues: Multi-Disciplinary Perspectives (ed. Mark Cartledge; Carlisle, UK: Paternoster, 2006), pp. 81-110. © Koninklijke Brill NV, Leiden, 2011

DOI 10.1163/174552510X526269

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

123

first and foremost as performance, then it is performance that is less concerned with the form of tongues-speech and more concerned with the context of its delivery. Helping me along the way are Martin Buber and Abraham Heschel, two Jewish philosophers who may not have had tongues-speech in mind, but whose insights into faith and spirituality offer to connect the ‘ineffable in us with the ineffable beyond us’. From There to Here As a third-generation Pentecostal, I am one of those who, it is frequently said, represents the beginning of the end of any religious movement that actively incorporates religious ecstasy as a significant part of its practices. Nonetheless, tongues-speech as ecstatic speech has been an important dimension of my prayer life since I became a card-carrying tongue speaker some 35 years ago. As a teenager, I understood that Pentecostals, at some point in time, need the ‘baptism’ with the accompanying evidence of speaking in tongues. I now spoke in tongues. I now had the baptism in the Spirit. Later in the mid 70s, while attending a Pentecostal Bible College, I was further ‘baptized’ into why God evidently chose this quirky gift to initialize believers into the baptism of the Holy Spirit. Evidently, God chose tongues as His ‘initial evidence’ of Spirit baptism because (1) they (tongues) constitute a visible symbol of spiritual reality; (2) they are uniformly recognizable by all cultures; (3) they reflect the personality of the Spirit; (4) they symbolize the Spirit’s complete control of the believer; (5) they reveal the Holy Spirit as the believer’s source of truth and utterance; (6) they signify the honor that God has placed upon human speech; and (7) they are a foretaste of heavenly speech.2 To this as students we could have added: ‘No tongues, no credentials’ for ministry upon graduating. Fortunately, that was not a pressure that concerned me; after all, I had my evidence before I went to college, and I really had no reason to question any of it. My tradition accepted this explanation; why shouldn’t I? However, upon graduation, it was not long before my own intellectual curiosity and the realities of pastoral ministry combined to raise some questions for which my previous college education offered little help. I spoke in tongues; now what? Or in my darker moments, I spoke in tongues; so what?

2 Carl Brumback, What Meaneth This? (Springfield: Gospel Publishing House, 1947), p. 235.

124

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

And in my quiet moments, I wondered what all good Pentecostals wonder at one time or another: whether tongues-speech could simply be learned behavior. Fortunately, Frank Macchia’s exploration into tongues-speech as an acoustic sacrament was a timely and crucial incursion into my own spiritual journey. In the same way a gothic cathedral says God is majestic, Macchia, drawing on the insights of anthropologist William Samarin and others, made the compelling case that tongues-speech says God is here.3 This revelation came on the heels of my freshly minted MA degree from Laval University, a French Catholic teaching institution. Not only did Macchia’s thought resonate with my new appreciation and understanding of catholic sacramental theology, but also for the first time it was a way for me to integrate tongues-speech into a richer holistic spiritual worldview that moved beyond the merely apologetic teachings of what I received in my theological training a decade earlier. It also had the additional benefit of introducing me to the Jewish philosophers Martin Buber and Abraham Heschel—two companions I have subsequently learned to cherish. While Macchia’s insights allayed many of my own questions, I confess that not all was resolved. Among other things, I still wrestled with making sense of what appeared to be a disparate functional biblical witness when it came to the subject of tongues-speech. Apparently, tongues-speech can operate missionally as a way of instant communication with peoples of different known language groups, and it can operate as a form of ecstatic prayer speech with God; of no known language group, it operates evidentially as a sign of Spirit baptism, and it can operate corporately as a source of prophetic inspiration when accompanied by the gift of interpretation. To be sure there is no shortage of systems by prominent Pentecostal theologians to explain such anomalies, but their explanations did little to allay my own nagging questions. Are we talking about different kinds of tongues? Is there a gradient scale of operation? Are Luke and Paul, the two primary protagonists on the subject of tongues-speech, simply addressing the same subject from different theological agendas?4 Or is their experience of tongues entirely two different entities? In addition to all of this, I still wondered silently about the divine nature of tongues-speech. Are the sounds emitted divinely imparted by God, or are they better classified as some sort of learned behavior? To be sure, if asked by

Frank Macchia, ‘Sighs too Deep for Words’, Journal of Pentecostal Theology 1 (1992), p. 53. I confess the earlier debates over this question among members of the SPS provided no end of memorable moments as a younger participant. 3 4

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

125

the uninformed, I knew the answer to these questions, but I was a long way from being satisfied with my own response. Into this personal journey, several years ago during the course of the 35th annual meeting of SPS, James Smith’s paper, ‘Tongues and Philosophy of Language: Conceptual Production at the Limits of Speech’, opened for me another personal avenue of investigation. Engaging, in particular, the philosophers J. L. Austin and John Searle, Smith considered the conceptual and ethical implications of tongues by viewing tongues-speech through the lens of speech-act theory, where the emphasis shifts from what tongues-speech ‘means’ to what tongues-speech ‘does’. In summary, Smith invited us to consider tongues as performative language that ‘arises out of resistance to given cultural norms and institutions’.5 In this case, perhaps the very inherent dissonance in glossolalic prayer says, in no uncertain terms, ‘all is not right’. In the words of Smith, it is an ‘act of resistance’ that, I might add, in corporate settings has the democratic additional benefit of stirring solidarity with others as they cry out to God to act on their behalf. From this perspective, we might even imagine that the anomaly of missionary tongues-speech in Acts 2, where the disciples speak in known foreign tongues but unknown to themselves, could be an act of resistance to a confused crowd that came to expect such authoritative boldness from those appropriately schooled—not from Galileans or anyone so uneducated. In this perspective, tongues-speech is more than just a prayer language of devotion; rather, it expresses a deep desire to beseech God when we do not know how to pray—a theme that resonated with Macchia but, in my judgment, moves in a different perhaps complimentary direction. If tongues-speech functions as an acoustic sacrament for Macchia, an action of grace bestowed from God to the devotee, tongues-speech, at least from the perspective of speech-act, appears to function as an acoustic icon where the emphasis is on a human initiation to transcend the ordinary and achieve something extraordinary. From a contemporary Pentecostal perspective, Smith’s hypothesis is bold and certainly not without controversy. And here my interest is only partially roused by Smith’s hypothesis of resistance, an emphasis in my judgment that has merit on its own; rather, it is on his conclusion that the content or, could we say, the form of tongues-speech may be irrelevant to its overall result. Does Smith not potentially blur the line between glossolalic speech as a gift of

5 James K.A. Smith, ‘Tongues and Philosophy of Language: Conceptual Production at the Limits of Speech’ (paper presented at the 35th Annual Meeting of the Society for Pentecostal Studies).

126

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

divine origin, a classical Pentecostal mainstay, and glossolalic speech as a tonal human piece de resistance reflecting ‘sighs too deep for words’? Or perhaps such a distinction is artificial? At this point, Smith appears to be in uncharted waters. In a departure from the speech-act theories of Austin and Searle, who reserve the application of speech-act for language that is rule-governed, Smith duly notes that not all forms of communication follow the conventional rules of language. Communication is more complex or nuanced than following a standard subject-verb-object agreement. However, his suggestion that the content of the tongues-speech is ‘basically irrelevant’6 raises the inevitable question, could not the effect of tongues-speech be the same through mimicry or divine impartation?7 With the right inspiration and the right accompanying elements in place,8 could not any incoherent ecstatic utterance produce the desired effect? By shifting the attention of tongues-speech from the illocutionary act to the perlocutionary effect of incoherent speech, Smith’s hypothesis certainly opens the door to such a possibility. For instance, could a Latin mass duly performed in the appropriate sacred place for congregants who do not speak Latin be a form of tongues-speech if it has the desired effect of initiating a certain communion with God that some might define as Spirit baptism? But before we widen the act of tongues-speech beyond supernatural gifting, it would be helpful to revisit the dialogue partners of Macchia, namely Abraham Heschel and Martin Buber, whose holistic approach to prayer quiets, in my opinion, the inherent tension between sacred and secular performance. Abraham Heschel, Quest for God: Studies in Prayer and Symbolism At first glance, Abraham Heschel seems like an unlikely partner to use with J.L. Austin and John Searle, whose inherent pragmatic approach to language 6 Unless of course we can claim the very incoherent nature of tongues-speech as an act of resistance is its content. 7 Or to rhetorically frame the question in another way, ‘Are children two years of age aware that they are learning a language through mimicry?’ If they understood this process of language acquisition, would it make any difference in achieving their results? Would they think any less of their acquired speaking skills? 8 J.L. Austin, one of the chief architects of speech act theory, is clear that for a performative phrase to have a happy result, ‘the particular persons and circumstances in a given case must be appropriate for the invocation of the particular procedure invoked’ (How to do things with Words [Cambridge: Harvard University Press, 1962], p. 34).

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

127

bristles against the mysticism of Heschel and his ineffable approach to God. About the only thing they seem to have in common is their preoccupation with speech. Nonetheless I believe a mild-mannered speech-act is not that far from Heschel. The key to understanding Heschel’s focus on prayer is found in the Jewish concept of Kavanah, or what he describes as ‘the yielding of the entire being to one goal, the gathering of the soul into focus’.9 For Heschel, the act of prayer itself trumps its motive and its content. Or to put it into a contemporary illustration, the salaries of professional athletes may be their impetus to play, but when true athletes are playing, the idea of their salaries is far from their mind. Winning is generally better than losing, but a player would rather be playing on a poor team than sidelined on a good team because of an injury. Athletes play for the love of playing, or so we mere mortals like to believe. And so in Heschel, the primary purpose of prayer is not to inform, or as in speech-act, perform; rather, it is to partake or indulge. And on this level, the language of prayer is often incidental to the act of praying. In fact, Heschel concludes there is little difference in essence between prayer as a stammer or prayer as eloquent discourse. But, should one be obligated to choose one over the other, Heschel states that simplicity is preferential over eloquence, given that the latter is often betrayed by twisting phrases, traps, and decoys that seduce the motives of the heart. Even ritual prayer may be preferred over eloquent speech, insomuch as the ritual itself can divert attention away from self and allow the devotee an opportunity to present herself unadulterated before God. Prayer for Heschel may be an ontological necessity as the way to God, not unlike a corridor in a hotel, but it suffers when it becomes the end in itself. What is at stake is the ability of the devotee in the act of prayer to ‘simplify self and to make God relevant to oneself ’.10 God is the great ineffable, concludes Heschel. And as such, God cannot be presumed upon as if someone thinks they can master a response to God. Accordingly, prayer is not a sermon delivered to God; it is ‘unbosoming oneself to God’,11 an action that is ‘guided by order and outburst, regularity and spontaneity, uniformity and individuality, law and freedom, a duty and a prerogative, empathy and self-expressions, insight and sensitivity, creed and faith,

9 Abraham Joshua Heschel, Quest for God: Studies in Prayer and Symbolism (New York: The Crossroad Pub, 1984), p. 12. 10 Heschel, Quest for God, p. 17. 11 Heschel, Quest for God, p. 17.

128

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

the word and that which is beyond words’.12 From this perspective, tonguesspeech as a kind of post-language13 has a great deal of potential insofar as it deflects attention away from the means in the quest to draw closer to God. In fact, Heschel is clear that we should not think of prayer as analogous to the give and take of human conversation. Prayer as Kavanah only makes us communicable to him as we pour out our hearts before him. But neither is prayer simply silence. Prayer engages words/sounds, but prayer is not found in the words themselves. To read Heschel and early Pentecostal literature on Spirit baptism, it may seem natural to make a connection between Jewish Kavanah and Pentecostal Spirit baptism. Both concepts describe a transcending state that climaxes in an ecstatic experience. In both cases, while words or sounds are the means into the ‘state of presence,’ the state is not defined by speech. It is a necessary passage that connects the two, but neither God nor humans are defined by it. From this perspective, one can understand why many Jewish worshippers who do not speak Hebrew may still insist that their Shabbat liturgy remain in the Hebrew language. The rhythm and cadence of the language, with its attachment to their long history, is in a unique position to lift the worshipper into the place of God. Heschel may even argue that such a possibility increases for those who cognitively do not understand the language. Of course, for classical Pentecostals, any notion that tongues-speech is less than a specific gifting from God is problematic. However, I suspect for Heschel that such a distinction would be artificial. All language is a gift from God. The ability of humans to communicate with each other and make themselves communicable with God is a gift whether the language follows known normative patterns of human speech or falls outside of them. Heschel is instructive when he recounts Moses’ encounter by God on Mount Sinai. For Heschel, this encounter is not a single one-time occurrence never to be repeated again; rather, it is a paradigm of both humanity’s need of God and God’s need of humanity. God, who is unwilling to be left alone, seeks to draw humanity by not only presenting himself but also equipping his 12 Heschel’s description of prayer as noted here certainly resonates with anyone who has participated in or witnessed a chorus of people employing tongues-speech in corporate prayer. To the observant, the relative incoherent nature of tongues-speech is balanced with a specific, coherent rhythm and structure. 13 I opt for the nomenclature post-language rather than either a-verbal or primitive to describe the ontology of tongues-speech. There is a layer of complexity in tongues-speech that may be overlooked in the other two descriptions where primitive implies an effort to recapture the 1st century usage and a-verbal implies a move away from language itself. By post-language I am suggesting an organic language that simply refuses to be categorized by rational discourse.

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

129

creation to respond in kind. And when we as his creation suffer ourselves to dwell with God, some prefer to wear a mask to conceal God’s emanating glory while others are content to blush when the ‘ineffable in us stands before the ineffable beyond us’.14 Could tongues-speech be a blush in the presence of God when the mask is removed and we are left chasing words? Of course, this bypasses the question of whether mundane words themselves can be a form of speaking in tongues if, in fact, they lift those speaking or hearing into an encounter with the presence of God. Again, the emphasis is not on the content of the language or its meaning but on what it is, an ontology that I suggest is only made known through what it does in very specific contexts. And what it does is function as an icon that potentially allows those seeking after God to go through language into an audible transcendent communion with God—a state we could perhaps entertain as Spiritbaptism.15 At this stage, two caveats are in order. First, for the language form to duly function in this way, users must agree upon a certain accepted conventional context and procedure. Is this not what Paul at least in part is addressing in 1 Corinthians 12–14 with his emphasis on order? Or to put it more bluntly, I could speak in tongues in this conference room right now, or more precisely I could utter some phonemes, but it would not be tongues-speech at least in terms that the Apostle Paul or Luke considered. The occasion is misplaced, and any perlocutionary effect would be unhappy to say the least.16

14 Abraham Heschel, Man is Not Alone: A Philosophy of Religion (New York: Harper and row, 1951), p. 91. 15 There is still room here to employ sacramental language in so far as understanding language as a grace conferred by God that allows humanity the opportunity to be communicable before God. 16 Of course, tongues-speech is not the exclusive domain of the religious. For aficionados of Pop Culture tongues-speech makes an appearance in the oft-time irreverent and always subversive DC comic series The Invisibles. See Grant Morrison, Jill Thompson, Dennis Cramer, The Invisibles: Arcadia part IV, No. 8 (New: York, DC Comics, April 1995). In an email exchange between Nicolas Greco, a friend of mine, and Grand Morrison, the author of The Invisibles, Morrison was asked why he introduced a ‘glossolalic speaking head’ into the series. Morrison responded, ‘I was doing a lot of speaking in tongues’ experiments in the early 90s and it occurred to me that the seemingly incoherent sounds of glossolalia could be seen as the voice of the ‘subconscious’ body mind and could be interpreted by the unconscious minds of others–everyone hears different words in a string of glossolalia. Everyone makes their own interpretation of such ‘inspired’ vocalizations, based on body language cues, intonation and the mishearing of phonemes according to personal bias. We all hear what we need to hear in a glossolalic exchange, making it a truly international language. Or so it seemed to me at the time’. While Morrison’s intuitive connection between unconscious communication and glossolalia is interesting, his

130

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

I feel the same way when I travel and worship in an assembly where tonguesspeech as an ecstatic unknown language would seem like an intrusion. I do not think any less of these people, and if it is a healthy congregation, it does not take long before I pick up the communicable signals that the congregation uses that contribute a similar perlocutionary effect whereby the congregants come into contact with God as he passes by them. Second, such a position necessitates rethinking the gift of interpretation in relation to tongues when used in a corporate setting as outlined in 1 Corinthians 12–14. In traditional Pentecostal theology, the gift of interpretation is understood as just that—a gift that enables someone to translate the audible sounds into a known discernable language for corporate edification.17 If, however, as in a performative icon, the meaning is not found in the actual sounds, an interpretative act is still required to translate what God is doing and subsequently saying through this interruption in the liturgical order. From Here to There: Martin Buber Martin Buber adds what could be another variable into this investigation. Buber has much to say about language and spirit. In fact, accordingly, the spirit is found in language. For Buber, the issue is not between learned language or what we might call ecstatic language, a difference he does not seem to make, rather it is a question of posture or positioning. Where does one stand in relation to language? Does one stand in language or outside of language? For Buber, one stands in language in the measure one receives from God as in an I-Thou correspondence.18 One stands outside of language in the measure one uses language to objectify the other or carry out the stuff of daily life. In the former case, language/Spirit is the doer and we are the done unto, whereas in the later case, we are the doer and language is the done unto. Writes Buber,

conclusion that we all ‘hear what we need to hear’ fails to take into consideration context. I argue it is precisely the context behind the occasion of tongues-speech that potentially transforms the ‘incoherent sounds’ into a transcendent spiritual event worth experiencing. 17 This can have interesting and dramatic results. In one church I attended, an English visitor in a French congregation sang an interpretative message that someone else had given through tongues-speech. As a result the presiding pastor sang the interpretation of the interpretation in native French for the benefit of the congregation. Presumably if what was required was an exact translation of specific words, God could have gotten it right the first time. 18 Martin Buber, I and Thou (New York: Charles Scribner’s Sons, 1970).

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

131

Man speaks in many tongues of language, of art, of action—but the spirit is one; it is response to the You that appears from the mystery and addresses us from the mystery. Spirit is word. And even as verbal speech may first become word in the brain of man and then become sound in his throat, although both are merely refractions of the true event because in truth language does not reside in man but man stands in language and speaks out of it—so it is with all words, all spirit… Spirit is not like the blood that circulates in you but like the air in which you breath…Man lives in the spirit when he is able to respond to his You.19

And here we take another cue from Buber, who suggests that whenever we find ourselves breathing the Spirit or living in the Spirit and try to objectify the moment through words, the moment is largely lost, and God who was our subject, becomes a he or a she or an it, in any case an object. Again, we might conclude that since all language is inherently sacred in so much as it is a gift from God, even mundane language can transcend the ordinary and bring people into a special place of communion with God. But second, from a Pentecostal vantage point, Buber’s observation about the relation between spirit and language suggests that tongues-speech as an ecstatic post-language discourse has a potential perlocutionary effect on the speaker by avoiding any objectification of God and prolonging the audience with God. To use a simple illustration, a couple under a starry sky may be lost in each other’s wonder. They may respond in silence, in an embrace, and they may even use words— but unless they have objectified the moment, I suspect that to anyone who overhears the words it will be as if they are speaking in tongues. Furthermore, this elocution in ‘tongues’ has an additional perlocutionary effect on anyone listening. In the case of the couple lost in love, it can encourage and challenge onlookers to review their own love relationships, recalling similar moments in their own lifetimes. The moment also points forward as individuals anticipate experiencing their own future I-Thou moments. In any event, the emphasis is not found in the words but in the motives and experience that lie beyond the words. But it should also be noted that any derivative edification from the moment cannot replace the experience of the moment itself, any more than a sentimental Hallmark card can accurately convey the experience of being in love. However, with the right variables in place the witness of such an authentic display of attachment can function as an acoustic icon that has the potential of occasioning the event in the life of the witness. In the end, we could conclude that such speech is not rehearsed or learned speech; rather, it is a suspension of rational discourse because the heart is limited and sometimes deceitful. It is limited in its capacity to find the right words to express the ineffable, and it is sometimes deceitful in that, given a 19

Buber, I and Thou, p. 89.

132

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

chance, it chooses words that reflect self-interests that invariably betray the moment. In his theological autobiography, Bound and Free, Professor Emeritus Douglas John Hall perceptively notes, ‘in theology one has to keep talking, because otherwise somebody will believe your last sentence’.20 And as I began this short interrogation into tongues-speech with some musings on my own personal journey, I realize this is simply part of a ceaseless conversation of my own, as I risk enlarging the borders of what we define as tongues-speech. At stake is not the content of tongues-speech nor its ‘supernatural’ character, as all language finds its origin in God. At stake is its unique performative essence that, when properly executed, allows worshippers to immerse themselves in the Spirit of God. Tongues and Language: Renewing the Linguistic Study of Glossolalia Matthew Wolf Introduction Are tongues ‘language’? For most people who speak in tongues, the answer is a simple yes. Some tongues-speakers might qualify their answer by suggesting that they may be heavenly languages; others might profess not to know.21 Yet, for the most part, linguists have concluded that Pentecostal tongues (usually called glossolalia in linguistic research) are not language. Summary statements such as the following are common in the literature: Glossolalia is of greater social and psychological than linguistic interest, since the vocalizations do not constitute language.22 When we comprehend what language is, we must conclude that no glossa, no matter how well constructed, is a specimen of human language.23 In spite of superficial similarities, glossolalia is fundamentally not language.24 20 Douglas John Hall, Bound and Free: A Theologian’s Journey (Minneapolis: Fortress Press, 2005), p. 24. 21 William J. Samarin, ‘Making Sense of Glossolalic Nonsense’, Social Research 46.1 (Spring 1979), p. 97. 22 Richard W. Howell and Harold J. Vetter, Language in Behavior (New York: Human Sciences Press, 2nd edn, 1985), p. 206. 23 William J. Samarin, Tongues of Men and Angels: The Religious Language of Pentecostalism (New York: Macmillan, 1972), p. 127. 24 Samarin, Tongues, p. 227.

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

133

Tongues-speakers may be surprised and disturbed by these conclusions (particularly when they are taken up by theological opponents to ‘disprove’ the divine origin of tongues). They may be repulsed from the linguistic study of tongues, either despising it as ‘godless’ or fearing its potential to undermine faith. On the other hand, the long-standing drought of linguistic research on glossolalia indicates that linguists have also retreated from the field, concluding that, as ‘non-language’, glossolalia has nothing more to contribute to our understanding of language. This paper argues that, on the contrary, practitioners of tongues and linguists alike should be interested in renewing the linguistic study of glossolalia. It is an area of inquiry from which linguists have many things to learn and tongues-speakers have nothing to fear. I will first seek to clear away two barriers to reopening the inquiry: (1) the desire to ‘test the supernatural’ with science and (2) the construal of ‘language’ as a Platonic category. Then I offer two research questions that may allow linguistic inquiry on tongues to move forward. Finally, I attempt to apply the arguments of this paper to the question ‘Are tongues language?’ as it is posed in theological discussions. Testing the Supernatural In one of the first linguistic studies of tongues, James Jaquith notes that his informant had ‘(1) gone to a Berlitz office and (2) approached “foreign looking” university students’ in the weeks before meeting Jaquith, ‘attempting in both cases to identify the “language” of his verbalizations’.25 It is not clear whether the informant turned next to Jaquith, a linguist, with the same goal, but his desire to empirically verify his spiritual gift is evident. This impulse to ‘test the supernatural’ is understandable; both the informant and the researcher no doubt were curious to know whether they were witnessing a miracle beyond human explanation. However, the desire to prove the miraculous should not be what drives Pentecostals’ engagement in the sciences. In an article arguing for Pentecostal engagement with modern science, Amos Yong interacts with studies in the cognitive sciences that describe Pentecostal experiences in terms of achieving a trance or alternative states of consciousness.26 Yong is ‘less interested in a kind of defensive-minded 25 James R. Jaquith, ‘Toward a Typology of Formal Communicative Behaviors: Glossolalia’, Anthropological Linguistics 9.8 (1967), p. 8. 26 Amos Yong, ‘Academic Glossolalia? Pentecostal Scholarship, Multi-disciplinarity, and the Science–Religion Conversation’, Journal of Pentecostal Theology 14.1 (2005), p. 61-80.

134

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

exegesis…or a kind of apologetic response’ to such studies. Instead, he welcomes the fact that ‘understanding the workings of the Holy Spirit in and through the neural processes of the brain will enable Pentecostals to get beyond the false dichotomy of naturalism versus supernaturalism’.27 Along the same lines, Malony and Lovekin believe that science has established ‘that the speech of glossolalics cannot be demonstrated to be a language’. However, they endorse C. S. Lewis’s notion that tongues represent ‘a lower structure…penetrated with higher meaning’. This is a premise which cannot be proved or disproved by scientific validation. They critique the presupposition (shared by many researchers and tongues-speakers alike) that a phenomenon like glossolalia is only divine if its immediate causes cannot be accounted for.28 These arguments should shape tongues-speakers’ expectations about what research on glossolalia is for and what it will discover. Empirical inquiry is devoted to the search for such ‘immediate causes’ as neural, mental, and social activity; those who believe in a God who created and controls the universe should not be embarrassed when empirical inquiry finds immediate causes. They should not be surprised to find the phenomenon mediated through ‘natural’ structures. (Having said that, believers in a God whose ways are beyond our ways should not be surprised if a phenomenon turns out to be seemingly unaccountable by ‘natural’ structures.) The Definition of ‘Language’ The notion that ‘language’ is a well-defined and broadly agreed-upon category often appears in discussions of glossolalia. William Samarin, who has published more on glossolalia than any other linguist, insists that whatever the glossolalist’s perceptions about tongues may be, ‘We [i.e. linguists] know enough to declare what is and what is not language’.29 Kildahl states that ‘linguistic scholars work with precise definitions of what constitutes a natural human language. Glossolalia fails to meet the criteria of these definitions’.30

Yong, ‘Academic Glossolalia?’, pp. 72, 68. H. Newton Malony and A. Adams Lovekin, Glossolalia: Behavioral Science Perspectives on Speaking in Tongues (Oxford: Oxford University Press, 1985), pp. 252-53. 29 Samarin, Tongues, p. 234. 30 John P. Kildahl, The Psychology of Speaking in Tongues (New York: Harper & Row, 1972), p. 47. 27 28

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

135

The most popular source given for these criteria is an article titled ‘The Problem of Universals in Language’ by Charles Hockett.31 It is worth noting that Hockett’s intention was not to set up a canon by which individual phenomena can be judged to be language or non-language. He writes, ‘If the next “language” on which information becomes available were to lack some feature we have believed universal, we could deny that it was a language and thus save the generalization’, but this would be mere ‘triviality’.32 His sixteen universals are inductive generalizations, ‘hypotheses to be tested as new empirical information becomes available’.33 Subsequent testing has shown some of his proposals are not universal, as he acknowledges in the second edition.34 Hockett is not troubled by this, since even when a language lacks some feature, that lack is ‘a typological fact of importance’.35 Given Hockett’s insistence on induction, one might have expected linguists to approach glossolalia as a potentially illuminating exception to his universals. To do so would not be to reject the value of the universals, but to recognize the limits of generalization. This is the kind of approach taken with sign languages, for example. The preference among human languages for the vocalauditory system over against hand signals is clearly nontrivial. Yet this nearuniversal does not negate the languageness of American Sign Language.36 Unfortunately, most researchers on glossolalia have taken Hockett’s universals as a list of necessary criteria, and in doing so, turned ‘language’ into a Platonic category.37 Givón explains that ‘categories of meaning within the Platonist tradition are clean and discrete, with no possible hedges, ambiguities or gradations…. Individuals can be either members or non-members in a category—but never part-members along a continuum’.38

31 Cited, for example, by Chilton, Goodman, Kildahl, Malony and Lovekin, Samarin, and Poythress; Jaquith cites an earlier article by Hockett. 32 Charles F. Hockett, ‘The Problem of Universals in Language’, in Universals of Language (ed. Joseph H. Greenberg; Cambridge, MA: M. I. T. Press, 2nd edn, 1966), p. 3. 33 Hockett, ‘Problem of Universals’, p. 2. 34 Hockett, ‘Problem of Universals’, p. 1. For example, Hockett included the vocal-auditory channel among universal design features, which excluded signed languages such as American Sign Language. ASL and other sign systems are now acknowledged as languages. 35 Hockett, ‘Problem of Universals’, p. 4. 36 See Edward S. Klima and Ursula Bellugi, The Signs of Language (Cambridge: Harvard University Press 1979). 37 Michael T. Motley, ‘A Linguistic Analysis of Glossolalia: Evidence of Unique Psycholinguistic Processing’, Communication Quarterly 30.1 (Winter 1981), p. 18-27. 38 Talmy Givón, Syntax: A Functional-Typological Introduction. Vol. I (Philadelphia: John Benjamins, 2nd edn, 2001), p. 29.

136

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

Samarin’s treatment of glossolalia conforms precisely to a Platonic understanding of the category ‘language’: ‘All languages ever known to man share fundamental features. Having defined “language”, we use this canon to decide whether glossolalia is to be admitted as one of its manifestations’.39 Samarin argues that glossolalia is missing five of Hockett’s sixteen universals: ‘semanticity, arbitrariness, displacement, prevarication, and reflexiveness’.40 This list sounds impressive, but when one reads Hockett’s definitions of these features, it is far from clear that glossolalia lacks all of them. When people pray in tongues that someone would receive the gift of tongues, they are demonstrating reflexiveness (speaking about language). When they pray in tongues for an absent person, they demonstrate displacement (referring to ‘things remote in time and space’41). Arbitrariness is simply the principle that there is no necessary connection between individual sounds and meaning. There seems no reason to propose that there is a necessary connection between glossolalic sounds and a certain meaning (as Samarin seems to admit in a later publication).42 Prevarication (the ability to lie) is an entailment of semanticity,43 and semanticity is the heart of the difference between glossolalia and a prototypical language. Hockett summarizes it with the proposition that ‘linguistic forms have denotations’.44 Although in a sense glossolalic utterances have semantics (they are ‘about’ something), they do not have semanticity: No correspondence has been found between glossolalic ‘words’ and particular denotations.45 Here we have an acknowledged divergence of glossolalia from Hockett’s universals. This divergence is often cited as the fundamental reason why glossolalia cannot be a natural language.46

Samarin, Tongues, p. 119. Malony and Lovekin, Glossolalia, 37. They cite William J. Samarin, ‘The linguisticality of glossolalia’, Hartford Quarterly 8.4 (1968), p. 49-75. 41 Hockett, ‘Problem of Universals’, p. 11. 42 Samarin, Tongues, p. 120. 43 Hockett, ‘Problem of Universals’, p. 12: ‘Without semanticity, a message cannot be tested for meaningfulness and validity’. 44 Hockett, ‘Problem of Universals’, p. 10. 45 Samarin, Tongues, p. 122. 46 E.g. E. Mansell Pattison, ‘Behavior Science Research on the Nature of Glossolalia’, Journal of the American Scientific Affiliation 20 (Sept 1968), p. 79. Also Samarin, Tongues, p. 128. Poythress (‘Linguistic and Sociological Analyses of Modern Tongues-Speaking: Their Contributions and Limitations’, in Watson E. Mills (ed.), Speaking in Tongues: A Guide to Research on Glossolalia [Grand Rapids, MI: Eerdmans, 1986], p. 478) makes ‘carrying information’ the most important factor, but he concludes that glossolalia does carry information through suprasegmentals (such as intonation) and even by lexical association. 39 40

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154 Platonic category: Everything is either a member or a nonmember.

Prototype–marginal category: Members cluster near the prototype, but some members are marginal.

137

Wittgensteinian category: Everything lies along a continuum.

Figure 1: Platonic, Prototype–marginal, and Wittgensteinian categories. Adapted from Givón, 2001.

Malony and Lovekin suggest that ‘it may well be that Hockett’s (1963) universals are not inclusive enough and that glossolalia is, indeed, a language in a different sense of the word’.47 Hockett’s commitment to induction invites such suggestions; and indeed, the erection and toppling of universals is a constant dynamic in linguistics.48 Rather than finding fault with Hockett’s list, however, I would take a slightly different approach. Inspired by Hockett’s principle that the absence of a near-universal is a ‘fact of significance’, I propose that we understand glossolalia as a marginal member of the category ‘language’. This understanding requires abandoning a Platonic definition of language, which admits no marginal members. But what do we put in its place? Wittgenstein suggested that rather than sharing essential qualities (as in Platonism), members of a category are related along a continuum of ‘family resemblance’.49 Givón evaluates Platonic and Wittgensteinian categories and proposes an ‘empirical middle ground’.50 This position mediates between Platonism’s discrete boundaries and Wittgenstein’s obliteration of boundaries; it imagines a prototypical member at the center of a category, with some members clustering around the prototype and others lying further out toward the margins. Malony and Lovekin, Glossolalia, p. 38. See ‘The Interpreter’ in the April 16, 2007 issue of The New Yorker for a fascinating account of an Amazonian language that may topple the supposedly fundamental feature of Universal Grammar. 49 Ludwig Wittgenstein, Philosophical Investigations (New York: Macmillan, 1953). Quoted in Givón, Syntax, p. 30. 50 Givón, Syntax, p. 31. He attributes the formulation of the middle ground to Eleanor Rosch, ‘On the internal structure of perceptual and semantic categories’, in T. E. Moore (ed.), Cognitive Development and the Acquisition of Language (New York: Academic Press, 1973); and idem, ‘Cognitive representation of semantic categories’, Journal of Experimental Psychology: General 104 (1975). 47 48

138

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

For our purposes, the most important characteristic of this system is that it allows graded membership: ‘The most prototypical member of a category is the one displaying the largest number of criterial features. But other members may display fewer features and still be members’.51 Givón’s preference for this position is not merely a philosophical impulse to split the difference between Plato and Wittgenstein. It is motivated by evidence from cognitive psychology that this is how the mind actually uses categories.52 As Givón writes, prototypes and graded membership account for two features of mental processing: (1) the rapid processing of most information by means of stereotyping (hence the prototype at the center); and (2) fine discrimination of ambiguous, marginal information (hence the shading away of members toward the margins).53 With this understanding, the discussion in the literature about how the linguistically naïve could be ‘fooled’ into thinking of glossolalia as language becomes moot.54 It is no mystery that ‘speakers and hearers’ of glossolalia ‘have assumed that what they heard was a connected sequence of sound from some language’,55 and they should not be thought of as ignorant for assuming so. In typical cognitive processing, the category ‘language’ no doubt has a speaker’s perceptions of his or her own language as the central prototype, with perceptions of other known languages and communicative phenomena (e.g. sign language, singing) shading away toward the margins. Glossolalia is intuitively classed as a member of this category, although a marginal one. A Better Research Question: How marginal is glossolalia? If a Platonic definition of language is abandoned, the question ‘Are tongues language?’ cannot be answered with a simple yes or no. A more nuanced question, one that invites thorough research into glossolalia, is ‘How marginal is glossolalia from prototypical language?’ One well-established way in which it is marginal is its lack of word–meaning correspondences (semanticity), but this is only one feature of the prototype. Givón, Syntax, p. 32. Ray Jackendoff, Foundations of Language: Brain, Meaning, Grammar, Evolution (Oxford: Oxford University Press, 2002), p. 25-26. 53 Givón, Syntax, pp. 33-34. 54 See Samarin, Tongues, pp. 104, 108, 121, 233-234; Malony and Lovekin, Glossolalia, p. 34. 55 Malony and Lovekin, Glossolalia, p. 22. 51 52

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

139

Unfortunately, because researchers cannot identify word–meaning correspondences, the number of tools with which they can study glossolalia is greatly limited. Even the study of glossolalic phonology (speech sounds) is hindered by this absence, since individual sounds cannot be contrasted in minimal pairs.56 To illustrate, if both /tik/ and /tig/ occur in a language and mean different things, one can conclude that the language differentiates /k/ and /g/—they are both phonemes. If a glossolalic utterance contains the strings /balatik/ and /tigana/, however, there is no way to tell if /k/ and /g/ are independent phonemes or expressions of the same phoneme.57 Nonetheless, several studies of glossolalic speech have been undertaken. These studies constitute a rather small body of work, mostly dating from the 1960s and 1970s.58 In general, they suffer from some unfortunate deficits, including imprecise (or absent) transcription and a very limited pool of subjects (mostly monolingual English speakers). Another deficit that will be discussed in the next section is a lack of integration with larger theories of language generativity. Still, the findings of these studies shed some light on how far glossolalia departs from prototypical language. Below I summarize the findings regarding four commonly studied features of glossolalic sounds: sound inventory, intersample reliability of sounds, relative frequency of sounds, and syllables. a. Sound inventory (number and kinds of sounds) It has generally been found that glossolalic sounds are a subset of the phonemes the speaker already knows.59 Samarin suggests that someone speaking in tongues ‘tends to use what is common in his native language. … he maximizes what is already most frequent’.60 Wolfram and Samarin both conclude that non-English sounds are rarely present.61 The relatively smaller sound inventory is generally taken to be evidence that glossolalia is non-language. For example, Samarin says, ‘The total number 56 Paul Chilton, ‘The Sounds and Sound-changes of Pseudo-language: A Case Study’, Anthropological Linguistics 21.3 (Mar 1979), p. 126. Motley, ‘Linguistic Analysis’, p. 19. Samarin, Tongues, p. 81. 57 It is for this reason that this paper refers simply to the sounds, or phones, of glossolalia rather than to phonemes (similarly in Motley, ‘Linguistic Analysis’). 58 Malony and Lovekin (1985) have a very thorough bibliography, and not much linguistic research has appeared since. 59 Malony and Lovekin, Glossolalia, pp. 35-36; Jaquith, ‘Toward a Typology’, p. 5; Samarin, Tongues, p. 83; Poythress, ‘Linguistic and Sociological Analyses’, p. 474. 60 Samarin, Tongues, p. 83. 61 Malony and Lovekin, Glossolalia, p. 36.

140

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

of different sounds appears to be smaller than one finds in most languages’; he estimates the average number at about 30.62 But by contrast, Motley’s samples ‘contain as many phone types (~30) as most languages do phonemes (25– 40)’.63 Moreover, subsequent research has put the typical number of phonemes in natural languages at 20–37, with 21 being the mode.64 No subject in any study that I have read has demonstrated less than the modal number of sounds (21) in his or her glossolalia, so there is no basis to judge glossolalia as linguistically marginal on the basis of sound inventory size. Likewise, the fact that glossolalia generally exhibits the most common, easiest articulated speech sounds (called unmarked sounds by linguists65) is taken as evidence of its being non-language.66 This is strikingly counterintuitive, since natural languages also prefer the least marked sounds. Conformity to the principles of markedness is a sign that glossolalia is more language-like; it shows that the mental processes which govern prototypical language also govern glossolalia. A comparison can be made to language acquisition, where research has shown that even when imperfectly acquiring a new set of phonemes, a language learner’s speech is consistent with markedness.67 The published literature and my unpublished research68 indicate a similar consistency for glossolalia. It is interesting to note that in addition to his sixteen well-known universals, which describe the general structure and function of language, Hockett proposes a list of phonological universals. I have not seen these referenced in the literature. Yet samples of glossolalia seem to be entirely consistent with Hockett’s list: having more than one vowel; a tendency toward phonological symmetry, though with inevitable gaps;69 at least two manners of articulation (stops + some other manner); and at least two places of articulation for stops.70 Samarin, Tongues, pp. 124-125. Motley, ‘Linguistic Analysis’, p. 20. 64 Ian Maddieson, Patterns of Sounds (Cambridge: Cambridge University Press, 1984). 65 See Michael Kenstowicz, Phonology in Generative Grammar (Cambridge, Mass: Blackwell Publishers, 1996), p. 62-65. 66 Samarin, Tongues, p. 83; Poythress, ‘Linguistic and Sociological Analyses’, p. 474. 67 E. Broselow and D. Finer, ‘Parameter setting in second language phonology and syntax’, Second Language Research 7 (1991), p. 35-39; M. Young-Scholten, ‘Interlanguage and postlexical transfer’, in A. James and J. Leather (eds.), Second-language speech: Structure and process (Berlin: Mouton de Gruyter, 1997), p. 187-210. 68 Matthew Wolf, ‘Phonological Structures in Glossolalia’ (unpublished thesis, University of Minnesota—Twin Cities, 2003). 69 That is, languages tend not to have nasals at a point of articulation where they do not have stops. This is an example of preference for unmarked forms over marked forms. Motley’s samples show language-like symmetry. 70 Hockett, ‘Problem of Universals’, pp. 25-26. 62 63

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

141

One feature, sound change, must be studied diachronically, and glossolalia has rarely been studied in this fashion. Some authors state, however, that a person’s glossolalia exhibits sound change over time.71 Hockett’s very minimal standards for phonological universals indicate that any given glossolalic discourse would have to show an extremely small inventory of sounds to be judged outside the realm of linguistic normality—far smaller than has been attested in the literature. b. Intersample reliability of sounds Another measure of phonological normality is whether a speaker consistently uses the same inventory of sounds over multiple samples. Samarin states that a glossolalic sound inventory is not stable;72 Chilton finds that it is not stable even within a single sample.73 On the other hand, Wolfram74 and Motley75 give very high statistical correlations between samples from the same speaker. c. Relative frequency of sounds Samarin characterizes glossolalia as ‘overworking’ its most common sounds: ‘There is generally a sharper break between the most and least common sounds than in natural languages’.76 Motley confirms that his samples show more ‘overworking’ of their most common sounds than does English; however, they show less ‘overworking’ than two other natural languages.77 d. Syllables Glossolalia’s preference for unmarked syllables (of the type consonant–vowel and consonant–vowel–consonant) has often been noted.78 This conforms to markedness, although Samarin finds that these common syllable types are also

71 Felicitas D. Goodman, Speaking in Tongues: A Cross-cultural Study of Glossolalia (Chicago: University of Chicago Press, 1972), p. 96. 72 Samarin, Tongues, p. 73. 73 Chilton, ‘Sounds and Sound-changes’, p. 140. One cannot help but wonder if it is relevant that his sample is not of religious glossolalia. 74 Cited in Malony and Lovekin, Glossolalia, pp. 35-36. 75 Motley, ‘Linguistic Analysis’, p. 21. 76 Samarin, Tongues, p. 126. 77 Motley, ‘Linguistic Analysis’, pp. 21-22. 78 Goodman, Speaking in Tongues, p. 121. Poythress, ‘Linguistic and Sociological Analyses’, p. 474.

142

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

generally ‘overworked’.79 On the other hand, Motley’s samples contain many more syllable types, including consonant clusters. Even more interesting, he asserts that the formation of consonant clusters is not random but follows phonotactic rules.80 Although all the phones in his samples also exist in English, they were at times combined in ways that English does not allow.81 So how far does glossolalia diverge from prototypical language? In terms of speech sounds, it seems less marginal than most researchers have concluded. While some research suggests that intersample reliability and relative frequency of sounds are more marginal, sound inventories and syllables are prototypical. It may be that researchers’ familiarity with English, which exhibits a more diverse and marked sound inventory than many languages, has exaggerated the impression that glossolalia is marginal.82 A Neglected Research Question: How is glossolalia generated? Samarin asserts that glossolalia ‘starts in the brain’,83 and I have argued that tongues-speakers should not be surprised if the brain at least mediates the phenomenon. But this assertion highlights a neglected research question: How can linguistic theory account for the production of glossolalia at all? In standard models of the mental production of language, the structural foundations of an utterance are its semantics84 and syntax.85 That is, the structure is determined by the words chosen for the utterance and the grammatical combination of those words into phrases and sentences. Phonology is derived from syntax—it is merely a surface manifestation of the syntactical deep structures. Within such a system, even the idea of glossolalia seems absurd— how can there be a surface structure (the sounds) with no deep structure behind it? A satisfactory answer has not been produced. Goodman was the first to address the linguistic production of glossolalia, and she seems to be the only researcher to speak in terms of deep structures. She hypothesizes: Samarin, Tongues, p. 126. Motley, ‘Linguistic Analysis’, p. 22: ‘Natural languages contain phonotactic rules which allow certain phoneme combinations and disallow others’. 81 Motley, ‘Linguistic Analysis’, p. 22. I found similar results with one sample in my own research. 82 Motley, ‘Linguistic Analysis’, p. 22. 83 Samarin, Tongues, p. 228. 84 Motley, ‘Linguistic Analysis’, p. 25. 85 Jackendoff, Foundations, pp. 107-111. 79 80

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

143

[Glossolalia] is an artifact of the [dissociative] mental state … an audible manifestation of the rhythmical discharges of this subcortical structure … In Chomskyan terms … glossolalia is not the surface structure of a linguistic, symbolic code, of a linguistic deep structure, but rather…an artifact of hyperarousal dissociation. Or, we might say that glossolalia has for its deep structure the hyperarousal dissociation.86

Goodman’s hypothesis has not fared well, largely because it is evident that glossolalia is often produced independently of an altered mental state.87 More importantly, Goodman makes something of a category mistake by identifying neural discharges as deep structures. Linguists since Chomsky have built evidence for a complex set of mental structures which govern language, but even the simplest of these structures cannot yet be correlated to specific neural activity.88 Yet Goodman’s theory leaps directly from neural discharges to vocalization. Does glossolalia bypass all these mental structures and spring directly from subcortical activity? To the extent that it demonstrates languagelike features (such as conformity to principles of markedness), such a theory seems unlikely. Rejecting Goodman’s hypothesis, Samarin posits a ‘glossolalia-process’ that he analogizes to the formation of pidgins. The process is characterized by ‘primitivization’ or ‘regression’89: that is, deviation ‘toward “simplicity”’.90 (This is apparently to be distinguished from the theory of psychological regression proposed by Oates, which Samarin rejects.91) When speaking in tongues, the glossolalist reactivates ‘instructions that have lain dormant since childhood’,92 which allow him to produce the echoic forms typical of glossolalia.93 Research does suggest that the production of glossolalia involves a move toward simplicity (or unmarkedness), although this hypothesis should be tested among populations with native languages unrelated to English. Such

Goodman, Speaking in Tongues, pp. 124, 151-152. Jaquith, ‘Toward a Typology’, p. 1; Kildahl, Psychology, p. 2; Samarin, Tongues, p. 79. 88 Jackendoff, Foundations, pp. 19, 21-23. 89 William J. Samarin, ‘Glossolalia as Regressive Speech’, Language and Speech 16.1 (1973), p. 81. 90 William J. Samarin, ‘Sociolinguistic vs. Neurophysiological Explanations for Glossolalia: Comment on Goodman’s Paper’, Journal for the Scientific Study of Religion 11.3 (Sept 1972), p. 294-95. 91 Samarin, Tongues, pp. 41-42. See Wayne E. Oates, ‘A Socio-Psychological Study of Glossolalia’ in Frank Stagg, E. Glenn Hinson, and Wayne E. Oates (eds.), Glossolalia: Tongue Speaking in Biblical, Historical, and Psychological Perspective (Nashville: Abingdon Press, 1967). 92 Samarin, Tongues, p. 228. 93 Samarin, ‘Glossolalia as Regressive Speech’, pp. 79-81. 86 87

144

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

populations should be easy to find, given the spread of Pentecostalism around the world. If Samarin’s hypothesis about reactivation of childlike phonology is correct, it also should be fairly easy to identify those ‘instructions’ involved, since child phonology is a well-researched subject. Another theory is Roman Jakobson’s sound symbolism.94 Jakobson finds in the patterns of glossolalia evidence that speech sounds have some symbolic value in the human psyche.95 It is certainly difficult to explain certain wide-spread features of glossolalia, such as the prevalence of the ‘nd’ consonant combination.96 Jakobson also points out a cross-cultural pattern in glossolalia and nonsense words that is even more striking: the progression of vowels from i (as in pizza) to a (as in father) to u (as in due). This progression seems to me as fundamental as the eight notes of an octave; my father’s glossolalia typically begins with a phrase like /bolikím bolikám bolikím bolichiú/. Exactly what these sounds are meant to symbolize, however, is unclear, and post-Saussurean linguists have had little patience for sound symbolism.97 Unfortunately, none of the hypotheses about the generative process of glossolalia have received much follow-up research. Motley analogizes glossolalia to ‘the aerodynamics of bumblebee flight—theoretically it should not occur’.98 Yet his challenge to linguists to account for the non-randomness of glossolalia has gone unanswered. When linguists have studied glossolalia at all in the last quarter century, they have generally focused on its function rather than its structure. Samarin exemplifies this trend; although he argues that glossolalia remains part of the linguist’s field of study,99 his interests shifted to how it functions in its social context,100 no doubt because he regards its structure as non-linguistic. More recently, David Hilborn has offered a highly instructive account of glossolalia from the perspective of pragmatics. Interestingly, his introductory survey of the research ends with Motley’s challenge: researchers have yet to account for glossolalic structure in the absence of semantics. But Hilborn then

Samarin interacts briefly with this theory in ‘Glossolalia as Regressive Speech’, p. 82. Roman Jakobson and Linda R. Waugh, The Sound Shape of Language (Bloomington: Indiana University Press, 1979), p. 181. 96 Jacobson and Waugh, Sound Shape of Language, p. 216. It seems that imitations/parodies of tongues like ‘I-come-in-a-Honda’ and ‘shandai’ have some empirical verification. 97 Jacobson and Waugh, Sound Shape of Language, p. 186. 98 Motley, ‘Linguistic Analysis’, p. 25. 99 Samarin, ‘Regressive Speech’, p. 77. 100 Samarin, ‘Making Sense’, pp. 88-105. 94 95

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

145

tackles a question that Motley never addresses, namely, how ‘tongues-speech might, in fact, communicate’.101 I mention this not to take anything away from Hilborn’s work, which illuminates a crucial part of the picture, but merely to point out that other crucial parts remain to be illuminated. The way forward in the task of accounting for glossolalic structure will benefit from interaction with the latest work in generative linguistics, particularly Ray Jackendoff ’s refinements of Chomskyan theory. As has been mentioned, in Chomsky’s models, the source of language generativity is syntax. Jackendoff critiques this ‘syntactocentrism’ and suggests instead that ‘language comprises a number of independent combinatorial systems’.102 Language as a whole requires that these independent tiers (for prosody, phonology, syntax, and semantics) interface with one another; but in some cases, certain tiers are left out. For example, a word like ouch has phonology and semantics but no syntax.103 These adjustments to Chomskyan theory open up the possibility of assimilating glossolalia into a theory of language—overcoming the problem of structure in the absence of semantics and syntax. Jackendoff comes within an inch of incorporating glossolalia when he tries to explain what independent activity on the phonological tier would look like. He refers to ‘stored pieces of phonology lacking both syntax and meaning, for example nonsense refrains … used to fill up metrical structure in songs and nursery rhymes’. He gives examples like ‘fiddle-de-dee, hey-diddle-diddle’, and ‘hickory-dickorydock’.104 A much more significant phenomenon, and a better piece of evidence for Jackendoff ’s theory, is glossolalia, which also lacks syntax and meaning. But unlike ‘hey-diddle-diddle’, glossolalia is generative (combinatorial) and not merely a fixed piece of phonology. Incidentally, Jackendoff also critiques the prevalence of sharp boundaries for linguistic categories, arguing along similar lines as Givón that indeterminacy is more neurologically realistic.105 Perhaps there is hope that the trend represented by Jackendoff will be ready to reopen the linguistic study of glossolalia.

101 David Hilborn, ‘Glossolalia as Communication: A Linguistic-Pragmatic Perspective’, in Mark J. Cartledge (ed.), Speaking in Tongues: Multi-disciplinary Perspectives (Bletchley, UK: Paternoster, 2006), p. 117. My italics. 102 Jackendoff, Foundations, pp. 107-111. 103 Jackendoff, Foundations, p. 131. 104 Jackendoff, Foundations, p. 132. 105 Jackendoff, Foundations, p. 25.

146

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

‘Language’ as a Theological Criterion It is unlikely that a great number of readers will be inspired by this paper to undertake linguistic research on tongues. However, the issues addressed in this paper are relevant to a broader audience than just linguists. The question ‘Are tongues language?’ crops up in theological discussions as well. If this paper has shown the shortcomings of that question and offered a better way to think about tongues in relation to language, then that should impact these theological discussions as well. Pentecostal theology stands by its assertion that contemporary tongues are to be identified with the New Testament phenomenon called glōssai in Greek106 and that both tongues and glōssai can appropriately be called language.107 On a popular level, tongues that are perceived to be more languagelike are regarded as more mature,108 but Pentecostals generally have felt little burden to assess the authenticity of tongues by the standard of languagelikeness. Some non-glossolalists, on the other hand, have raised natural language as the standard by which tongues should be judged. In his essay ‘Glossolalia in the New Testament’, Frank Stagg describes the speaking in tongues at Corinth as ‘unintelligible’ and representing ‘a sickness’. He then insists that ‘babbling, ancient or modern, is Corinthian and not Pentecostal’. The authentic languages of Pentecost, on the other hand, correlate with ‘preaching the gospel, winning people to Christ, close-knit fellowship (koinōnia), teaching, praying, and sacrificial and joyful living’.109 For Stagg, tongues must be natural languages or else they are unspiritual babbling. While Stagg seems to equate modern tongues with Corinthian glōssai (like most commentators on glossolalia110), Jerry Corbaley, a trustee of the Southern Baptist Convention’s International Mission Board, argues that glōssai in both Acts and 1 Corinthians implies natural languages. He has challenged Pentecostals and charismatics to submit their ‘prayer language’ to

E.g. Acts 2.4; 1 Cor. 14.2. Malony and Lovekin, Glossolalia, p. 22. 108 Samarin, ‘Making Sense’, p. 99. Max Turner, ‘Early Christian Experience and Theology of ‘Tongues’: A New Testament Perspective’, in Mark J. Cartledge (ed.), Speaking in Tongues: Multidisciplinary Perspectives (Bletchley, UK: Paternoster Press, 2006), p. 15. 109 Frank Stagg, ‘Glossolalia in the New Testament’, in Stagg, Hinson, and Oates, Glossolalia, p. 41. 110 E.g. Kildahl, Psychology, p. 13. Samarin, Tongues, p. 235. 106 107

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

147

empirical verification. For Corbaley, modern tongues are either verifiable natural languages or they are gibberish.111 In one respect, the theological approach of Stagg and Corbaley parallels the scientific approach taken by most of the linguistic researchers reviewed in this paper: the construction of ‘language’ as a Platonic category. Independent of whether linguists should reject a Platonic definition of language, as this paper has argued, it certainly seems inapt to use a Platonic definition in assessing the spiritual authenticity of tongues. To do so assumes that Luke and Paul (and the Holy Spirit?) have a twentieth-century linguist’s understanding of language, complete with Hockett’s sixteen universals. This paper cannot attempt to define the influence of Platonism on the New Testament,112 but it is clear enough from 1 Corinthians that Paul does not apply sharp, Platonic boundaries to the term glōssai. Turner argues that Paul understood the term ‘in the very obvious and commonplace sense of “language.”’ Furthermore: Like most modern Pentecostals/charismatics without linguistic expertise, he would simply assume that utterances that sounded like unknown languages (and were capable of charismatic ‘interpretation’) were languages (earthly or heavenly) failing strong contra-indicators (and what in his world could those be?!).113

Paul has no difficulty using this neutral term114 to describe the phenomena at Corinth, even though he is aware that it lacks semanticity (1 Cor. 14.2, 11). Despite its unintelligibility, he does not reject the Corinthian phenomenon as non-language, but points out its marginality: Unlike ‘revelation or knowledge or prophecy or teaching’ (14.6, NRSV), it is not ‘readily recognizable, clear, distinct’115 (14.9). It fails to carry information, like a battle trumpet playing ‘an indistinct sound’ (14.8, NRSV). Paul’s assignment of such phenomena to a category like glōssai is not naïve or ignorant; as this paper has argued, its use of prototypical and marginal membership makes perfect sense from a cognitive perspective.

111 At the time of writing, Corbaley’s website dedicated to the topic, www.sbcglossolalia .blogspot.com, is no longer maintained. 112 See the article ‘Philosophy’ and its bibliography in Gerald F. Hawthorne, Ralph P. Martin, and Daniel G. Reid, Dictionary of Paul and His Letters (Downers Grove, Ill: InterVarsity Press, 1993). 113 Turner, ‘Early Christian Experience’, pp. 16-17. 114 Glōssai seems to have only become a semi-technical term for glossolalia after the NT, pace BDAG pp. 201-202. See Turner, ‘Early Christian Experience’, pp. 12-14. 115 BDAG’s definition of eusēmos, p. 413.

148

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

Because the most detailed interaction with tongues in the New Testament is in a letter to a spiritually immature church, opponents of tongues have often associated the phenomenon with spiritual immaturity.116 One can imagine this as a backlash against some who hold out the Spirit’s gift of tongues as a sign of their own spiritual advancement. What both sides of such an argument miss, however, is that neither the presence nor absence of tongues correlates with maturity. Tongues exist among the mature (Paul; 1 Cor. 14.18) and the immature (the Corinthians; 14.20); the distinguishing sign of maturity is love, without which even normal language becomes mere noise (13.1). So Paul is not interested in ‘language’ as a test of spiritual authenticity. He does not seek to verify whether the tongues at Corinth sound enough like language, or can be identified with specific languages. The proof of the Corinthians’ being ‘infants in Christ’ is their ‘jealousy and quarrelling’ (3.14); they will grow up not by abandoning tongues (14.39), but by exercising all of their gifts in love to build each other up (14.26).

Conclusion Are tongues language? This paper argues that the contradictory answers to that question can be resolved without doing injustice to scientific inquiry or the Pentecostal experience of tongues. The resolution starts by recognizing the superiority of prototype–marginal categories over Platonic categories. This acknowledgement replaces the opening question with a more fruitful one: ‘How marginal is glossolalia from prototypical language?’ This change of questions is not a religiously motivated attempt to ‘rescue’ tongues from scientific study. Rather, it is consistent with the scientific study of language and how the mind produces language. In terms of speech sounds, glossolalia is less marginal than the literature suggests, indicating that glossolalia is produced by at least some of the same mental systems that produce prototypical language. In the present state of research, glossolalia could be called a non-semantic or non-grammatical phenomenon,117 but descriptors like ‘non-lingustic’ or ‘pseudo-language’ are inapt.

116 Stagg, Hinson, and Oates, Glossolalia, is a premier example of infantilizing those who speak in tongues. 117 As in Hilborn, ‘Glossolalia as Communication’, p. 116.

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

149

There is reason to hope that changes in the field of linguistics may reopen the languishing inquiry into glossolalia with fruitful results. An important research question will be how theories of language generativity can account for the production of glossolalia. The move away from a Platonic understanding of ‘language’ also undercuts the attempt by some to use the scientific verdict on tongues as proof of spiritual immaturity or illegitimacy. Such attempts are shown to ignore Paul’s more cognitively realistic perspective on language, and his most important criterion: Christian love. Naturalizing Glossolalia?: Prospects for a More Radically Pentecostal Ontology James K.A. Smith I would like to take up a theme that I find very provocatively rumbling in the papers by Holm and Wolf—two papers that I think make very important contributions to furthering our thinking about tongues-speech. So instead of a rather tedious point-by-point engagement, I’d like to point to a common theme that is suggested by both of them, and then briefly entertain what it might look like to pursue this further—again, as a means of further developing the depth of pentecostal reflection on glossolalia. Wolf: Tongues on the Brain Wolf ’s fascinating paper118 takes up a question that has long been put to tongues-speech: ‘Are tongues language?’ And he very helpfully suggests that while this might be a legitimate question, it is illegitimate to impose a ‘Platonic’ or universalist definition of the ‘essence’ of language, thereby excluding anything that doesn’t exactly match the criteria.119 Instead, Wolf adopts a more elastic, Wittgensteinian definition of language which, when stretched, can include glossolalia as a ‘marginal’ language. I find his argument and

118 Consider also recent work on tongues included in Mark Cartledge (ed.), Speaking in Tongues: Multi-Disciplinary Perspectives (Carlisle, UK: Paternoster, 2006). 119 Interestingly, the critique of ‘Platonic’ accounts of language here is almost an exact analogue of Derrida’s critique of Husserl’s rather ‘Platonic’ account of language, as developed in Derrida, Speech and Phenomena.

150

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

evidence very convincing in this regard. What I find intriguing is just what motivates the question. Why is it important to determine whether tongues are a ‘language?’ What’s at stake in tongues qualifying as a ‘language?’ And what would we lose if it failed the test? Here I find an interesting tension between the opening and the conclusion of the paper. In his introduction, Wolf poses the question whether tongues are language, and notes that many—including both non-Christian linguists and theological opponents—have concluded that tongues are not language, and that therefore ‘tongues are merely human and not the work of the Holy Spirit’ (p. 1). While this dichotomy is not Wolf ’s, it is in the air: a curious opposition between tongues qualifying as ‘a language’ and tongues being ‘merely human’. Or, to state the implied inverse, if tongues do qualify as ‘a language’, then that would be proof that they are ‘the work of the Holy Spirit’. Those are both curious correlations: on the one hand, if they’re not language, they are human; on the other hand, if they are language, they’re somehow not ‘human.’ But by the time we get to the end of the paper, Wolf has convinced us that tongues do at least qualify as a ‘marginal’ member of the language family (kind of the crazy uncle of the family!), precisely because tongues exhibits the kinds of linguistic features that characterize natural languages, including how language is generated by the brain. Thus Wolf asks, ‘How can linguistic theory account for glossolalia at all? If glossolalia “starts in the brain”, how does the brain create these strings of sound which are language-like in terms of phonology and suprasegmentals, but cannot be parsed into morphemes and syntactic structures’ (pp. 11-12)? In reply, he invokes Goodman’s theory of ‘hyperarousal dissociation’—which I had always thought was just the sort of thing that got A/G pastors defrocked! But it turns out this actually accounts for tongues-speech! (I could say more here but will resist the temptation.) What I find exciting and intriguing about Wolf ’s conclusions is what, I suspect, many pentecostals would find worrisome and disturbing—namely, the suggestion that tongues bubbles up from the brain just like (normal) language. In other words, by answering the question, ‘Are tongues language?’, in the positive, Wolf at the same time makes the case that tongues are human. As he puts it, linguistics help us to appreciate ‘how the mind produces language’, and ‘that glossolalia is produced by at least some of the same mental systems that produce prototypical language’ (p. 18). In short, it seems to me that Wolf is suggesting that tongues are ‘natural’ and ‘human.’ Given the dichotomy in his introduction, however, must we therefore conclude that tongues are not the work of the Holy Spirit?

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

151

Learning to Pray: Holm Enter Randall Holm.120 What a delightful paper—Pentecostal theology as it should be written, laced with the poetry of testimony and an authentic concern for lived discipleship. (As we pentecostal theologians get more seats at the big theological table, so to speak, may we not abandon this unique genre that strikes me as a uniquely pentecostal style of theology.) I appreciate Holm’s honest wrestling with the issues. (I have my own familiarity with the ‘No tongues no credentials’ axiom, having been turned down for A/G credentials on just this point. I took it as the Spirit’s confirmation that I was destined to be a small-‘p’ pentecostal!) Holm also neatly tracks the development of theological and philosophical thinking about tongues up to the present, including Macchia’s reframing of tongues in sacramental terms, and my own attempt to shift the lens to one of doing and performance rather than meaning and communication. Bringing these together as catalysts, the fruit of the ensuing ‘reaction’ is Holm’s wonderful metaphor: tongues as an ‘acoustic icon’ (p. 4). I think he takes the conversation to the next level by drawing on Heschel’s and Buber’s theologies of prayer and I can only commend his suggestions as original and productive advances. But as with Wolf, I’m once again intrigued by how this is framed. For instance, Holm articulates that lingering pentecostal/charismatic worry, ‘whether tongues-speech could simply be learned behavior’ (p. 2). I take it that this is a fear or worry because we assume that if tongues were a ‘learned behavior’, that would mean that they are ‘merely human’—and not a divine gift. Thus he also worries a little—or at least wonders aloud—about my model which threatens to blur the distinction. As he puts it, ‘Does Smith not potentially blur the line between glossolalic speech as a gift of divine origin, a classical Pentecostal mainstay, and glossolalic speech as a tonal human piece de resistance reflecting “sighs too deep for words”? Or perhaps such a distinction is artificial?’ (p. 4).

120 I should note that Randy has had more impact on me than he perhaps knows. I was a late convert to Christian faith (the day after my 18th-birthday) and made my way to Pentecost a couple of years after that, in the bosom of Bethel Pentecostal Tabernacle in Stratford, Ontario. My pastor, Charlie Swartwood, thought very highly of some thoughtful teachers (and troublemakers!) at Eastern Pentecostal Bible College, including Ron Kydd, John Stephenson, and one Randall Holm. He put into my hands a stack of issues of the Eastern Journal of Practical Theology that included a number of essays by Randy. This was a huge part of my early formation as a budding pentecostal scholar and I am happy to here have an opportunity to repay the debt in a small way.

152

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

One can see how both Wolf and Holm bump up against a unique challenge for pentecostals: just how to think about tongues vis-à-vis ‘nature’. Or in other words, whether to think about tongues as either ‘human’ or ‘divine’. Naturalizing121 Glossolalia: A Modest Proposal I think it is a virtue of these two papers to pose exactly this question. And I think trying to answer it can be instructive beyond the narrow case of tonguesspeech. Let me begin with two confessions: First, I am not a classical Pentecostal precisely because I couldn’t buy the doctrine of initial evidence (well that, and some matters of A/G ecclesiology, but that’s another matter). So I have to admit that perhaps the stakes are not quite as high or charged for me as they might be for others. However, I would say I don’t buy the doctrine of initial evidence, not only because of exegetical issues, but also because of the ontology that it assumes. More about that in a moment. Second, when first making my way to Pentecost, so to speak, I remember going to the altar and praying like mad for the baptism of the Holy Spirit with the initial evidence of speaking in tongues. And I particularly remember old Jack, who always sat in the front row, coming over to me, wanting to help me in this quest, encouraging me to just begin stuttering a bit, to just let my tongue loosen a little, to begin babbling a few syllables to prepare the way, as it were, for the Holy Spirit. I have to tell you: at the time, I was aghast and disappointed at this. Coming to the altar with an expectation of being overwhelmed and bowled over, of being a passive recipient who would be inundated by a linguistic outpouring beyond my will and power, it was entirely disenchanting for me to hear this old saint encouraging me to, well, kind of ‘fake it’ (as it felt to me) in order to get started. But I would say that I have come to a new affirmation of what old Jack was up to, even if Jack might still be working with an ontology I find suspect (of course, Jack would be quite surprised to know he even had an ontology, but you know what I mean). In short, I think we do well to affirm tongues-speech as ‘human’, without thereby concluding that it is somehow inauthentic. Let me explain. First, I don’t want to unilaterally deny the possibility that glossolalia could be a sort of divine gift that comes quite irrespective of learned linguistic 121

I am playing on Quine here just a little bit.

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

153

ability (e.g. I don’t at all rule out the possibility of miraculous linguistic abilities being given in missional contexts; indeed, it seems to me that Acts 2 is an instance like this). However, that said, I do think we should follow Wolf ’s and Holm’s underlying suggestions and affirm that tongues-speech is human and is a ‘learned behavior’. But at the same time, I don’t think we should therefore conclude that this is not divine or not a gift of the Holy Spirit. To do so would assume a distinction which I think Holm rightly suggests is artificial, viz., the assumption that the human and divine—or the natural and the supernatural—are somehow mutually exclusive. I think that in many cases tongues speech is learned behavior; but what we’re learning is a kind of speech of the Spirit. Or, to put it in other words, I think tongues-speech is often learned in very ‘natural’ ways—but what’s being learned is a speech that pushes against the limits of ‘the natural’. Why should we continue to work with an opposition between what’s ‘natural’ and what’s ‘of the Spirit’ when the Spirit is the Spirit that inhabits creation/nature? When I pray in tongues, I am engaging in a linguistic practice that expresses and resonates with the fact that I find myself on the very edge of speech because I’m being stretched beyond the habits of the natural. Tongues-speech is a paradoxical linguistic expression of the failure of speech; it is also the sort of utterance that displaces one from the position of being a confident, articulate master of the situation. In short, it reduces one to a babbling idiot, but that is precisely the virtue of the practice: it is a mode of linguistic humility. Sometimes I pray or sing in tongues because of an experience of untold delight and joy that exceeds the everyday lexicon I’ve been given, but which nonetheless calls out for shouts or song. At other times, I pray in tongues because the heartbreak of inhabiting such a broken, volatile word eludes any kind of coherent articulation, and yet I need to pour myself out to God (what else could we do?). For me, praying in tongues is a humbling, kenotic mode of prayer—a sort of linguistic prostration before God. I have learned to do this. When I first went to the altar, I imagined that tongues-speech would come upon me—that, trance-like, I would become a mere conduit for the Spirit’s utterance. I imagined that my own will, volition, and mental capacities—my own neurons and synapses—would somehow become mere instruments of Someone else. I expected that my ‘nature’ would be overwhelmed, almost possessed, by Someone supernatural who would take over direction of my neural pathways and vocal chords. But it seems to me that this is not very good theology, not even good Pentecostal theology, precisely because such a picture is rather anti-nature and works with a mistaken dichotomy between the natural and the so-called super-natural, between the ‘human’ and the Spirit. Even the Third Person of

154

R. Holm et al. / Journal of Pentecostal Theology 20 (2011) 122–154

the Trinity is an incarnational God, both condescending to and taking up the ‘nature’ which is at the same time the creation which he made.122 Why should we think the gift of tongues would operate in such a way? Do we think the gift of teaching or administration operates in this way? If I have the gift of teaching, don’t I marshal and direct all my so-called ‘natural’ abilities and talents in order to channel them and see them taken up in Spirit-filled teaching? Might we not think of tongues-speech in the same way? So I think we should wholeheartedly embrace tongues as ‘human’, as a ‘learned behavior’—but then refuse to thereby conclude that it is not ‘of the Spirit’, precisely because we should refuse the natural/supernatural bifurcation that informs such a mistaken conclusion. And while I’m affirming that tongues is often a learned, human practice, I don’t think this is at all disastrous for pentecostal spirituality (though it might not be helpful to the doctrine of initial evidence). In fact, I think it is important to note that this is a practice that I learned from Pentecostals. This is because pentecostal spirituality lives on this cusp of the natural/supernatural distinction that we struggle to both articulate and elude. Pentecostal worship is a school of the Spirit that teaches us the cadences of such a ‘marginal’ speech of resistance. And that remains a gift.

122 In this respect, we might draw a helpful analogy between the dynamics of tongues-speech and ‘inspiration’, though I can’t pursue this here. For a suggestive line, see Peter Enns, Inspiration and Incarnation.