University of West Sydney
The way questions of gender and sexuality play out in Rabindranath Tagore’s texts has long been on the radar. His oeuvre is regarded proto-feminist at times, while on other occasions Tagore is seen as a writer who stops on the threshold of actualizing the final breakthrough of his own subversive gender discourse. Santosh Chakraborti, in his book Studies in Tagore: Critical Essays, compares him to Thomas Hardy in this regard. Both Tagore and Hardy, according to Chakraborti, are marked by an ‘uneven feminism’ (Chakrabarti, S. (2004); p.6) wavering from one text to another. Sexual politics in Tagore’s work is an integral part of the larger socio-political backdrop of the early twentieth century. Niharranjan Ray, in his Rabindrashahityer Bhumika, suggests Tagore was influenced by the Western discourses regarding women’s struggle for social justice; Santosh Chakraborti makes the point that this feminism is informed by the twentieth-century ideals of individualism.
This article seeks to handle a neglected aspect of Tagore’s presentation of women and examine whether it can become a site for negotiating the complexities of his attitude towards them. Ranjana Ash, in the article ‘Introducing Tagore in Multicultural Education in Britain,’ passingly refers to this strand; ‘Nashtanirh’ [‘The Broken Nest’], serialized at the same time as Chokher Bali, embodies another quality of Tagore’s women characters: their determination to develop their creative gifts’ (Ash, R. (1989); p.151).
I would underline the trajectory of writing in this path of creative gifts; my point of departure being Ranjana Ash’s comment on Charu in ‘Nashtanirh’: ‘Literary talent is neither comfort nor compensation in such a situation’ (Ibid.). In the trajectory of feminine writing in Tagore, there is something more at stake than the ‘literary.’ This writing is not only literary but also a fundamental expression of feminine desire; and as an expression of feminine desire through language, this writing creates an opening into an order, potentially outside of patriarchy. This is where the expression may be both ‘comfort’ and ‘compensation.’ In her article ‘A Sentimental Education: Love and Marriage in The Home and The World,’ Supriya Chaudhuri makes a significant comment on Bimala: ‘Her desire for a lost unity or integrity of being which she associates with the past, with her mother’s life’ (Chaudhuri, S. (2003); p.50). She further observes that inasmuch as Bimala’s ‘emotional history … is predicated upon this sense of lack, it may be read as a history of desire’ (Ibid.; p.51). According to Supriya Chaudhuri, Bimala’s desire returns upon itself instead of being ‘directed towards husband or lover’ (Ibid.). If we generalize and expand this view we can propose that feminine desire tends to fall back on itself instead of being sustained by specific external objects—like the husband or the lover, each of whom are implicated in the patriarchal structure of relational positions. In my view the self-enclosed nature of feminine desire triggers a movement beyond the patriarchal arrangement. The writing, which expresses this desire, is unmistakably feminine and I would argue that it goes as far as to the liminal point of patriarchy. In this reading I will see both the self-enclosure and the opening of feminine desire in writing as it makes an effort to open itself to its sexual Other—namely the man.
In this article I will examine the figure of the writing woman and by extension the sphere of feminine writing as expressed in three of Tagore’s short stories: ‘Khata’ [‘The Exercise Book’], ‘Nashtanirh’ [‘The Broken Nest’] and ‘Streer Patra’ [‘The Letter from the Wife’]. In all three stories we encounter a number of women writers, and their writing is placed right against the sphere of male writing. This sphere of writing, in these stories, becomes a trope for exploring the man-woman relationship and involves an asymptotic dialogue of two absolutely different kinds of writing. The overlap of the textual and the sexual is extended from the thematic to a larger tropological level where parts of the text or even the whole text (as in ‘Streer Patra’) functions as an alternative feminine discourse of subversion.
In Tagore’s short stories the female characters often produce disturbing silences when they encounter the articulations of masculine desire. In ‘Postmaster,’ Ratan remains silent when the postmaster tells her that he will never return to Ulapur. Tagore fuses this moment of silence with the image of a leak on the thatched roof of the postmaster’s house. The image makes the lack in their relation explicit. I would claim that the silence of Tagore’s women is not submissive: it bores holes in man’s understanding. This silence is not a pristine outside to language but rather a break in discourse, itself gradually mutating into a discourse. Tagore contrasts man’s theoretical inclination with a woman’s non-theoretical bent of mind by juxtaposing the theorizations of the postmaster while leaving on boat with Ratan’s obsessive circulation around the postmaster’s house. When the man is busy philosophizing, a woman responds through her wordless affective action. The narrator comments on this action—’Kintu Rataner mone kono tatyer uday hoilo na’ [‘But no such theory emerged in Ratan’s mind’] (Tagore, R. (2002); p.17). The non-theoretical realm is like a blank page on which the postmaster can only try and force an explanation—‘Bodh kari tahar mone khin asha jagitechhilo, dadababu jadi phiriya ashe’ [‘Perhaps there was still a faint hope in her mind that dadababu might come back’] (Ibid.). This is a forced interpretation, projected from the male position. Encountering his own incomprehension of feminine desire, man tries to supplement the lack by supposing an interpretation. The interpretation in this particular story bases itself on feminine dependence and an idea of ‘sympathy’ that makes a victim of her.
In the French psychoanalyst Jacques Lacan’s teachings, the most crucial difference between man and woman is that while the former is wholly cut by the Symbolic, i.e. language, the woman is partially cut by the signifier. Though both men and women are, for Lacan, divided subjects qua language, the division is never a whole in the case of the woman, and as a result there is more of the Real, or that which resists linguistic symbolization, in her. This makes the ‘the woman’ somewhat problematic, and Lacan marks her non-generalizable singularity with the expression ‘not all’ or ‘not whole’ [‘pas tout’]. This is the discourse which culminates in the famous Lacanian maxim—‘There is no such thing as Woman; Woman with a capital W indicating the universal’ (Lacan, J. (1998); p.72). As Lacan would say there are women but no Woman: in his 20th seminar in 1972-73, On Feminine Sexuality: The Limits of Love and Knowledge, while talking about the function of the written, Lacan says:
A man is nothing but a signifier. A woman seeks out a man qua signifier. A man seeks out a woman qua – and this will strike you as odd – that which can only be situated through discourse, since, if what I claim is true – namely, that woman is not-whole – there is always something in her that escapes discourse (Ibid.; p.33).
Although there is more of the Real in a woman than there is in the man, the only way she can try and reach out to the man is through the signifier. This is what activates the field of feminine writing in its relation to the loved one. If there is female silence as dissidence in Tagore, there is a counter-discourse of female writing as well. Chandara’s enigmatic last word ‘maran’ [‘death’] before her capital punishment in the story ‘Shasti’ [‘Punishment’] is an instance of a supplementary woman-speak. When asked whether she wants to see her husband Chhidam, one last time before being put to death, Chandara’s articulation of the signifier ‘death’ collapses the idiomatic valence of this word as a rustic expression of disgust. The idiom is de-idiomatized and made literal. And it is this literality of the word ‘maran’ that escapes patriarchal comprehension. Uma in ‘Khata,’ Mrinal in ‘Streer Patra’ and Charulata in ‘Nashtanirh’ are all writers in their own right and it is their writing that contests patriarchal literary authority.
The first line of ‘Khata’ declares that little Uma, learning how to write, creates ‘upodrob’ [‘tremendous trouble’] (Tagore, R. (2002); p.163) for one and all in the family. She writes on the walls of the house—‘Jal pare, pata Nare’ [‘Rain patters, leaves flutter,’ in William Radice’s translation] (Ibid.). Finding a copy of The Secret Adventures of Haridas under the pillow of her brother’s wife, she writes ‘Kalo Jal, lal phul’ [‘Black water, red flower’] (Ibid.) on multiple pages of the book. On her father’s notebook of daily expenses, she inscribes—‘Lekhapara kare jei, garighora chare shei’ [‘He who learns to write/Drives a horse and cart,’ in Radice’s translation] (Ibid.). With the exception of the first of these examples, where Uma writes on a surface which is blank, albeit non-writable, in all the other cases, her writing is a supplement added to a pre-existing text, be it The Secret Adventures of Haridas or the ‘daily expense’ notebook of her father. This marks her writing straightaway with a rhetoric of intervention. Her scribbles on all the almanacs in the house almost end up obliterating the pre-existing text which has an intrinsically providential character. The other crucial detail about her childish writing is her use of non-sense verses which again opens up the non-theoretical realm. Her lines on her father’s notebook are counterpointed with his numerical figures and the materialist thesis of education proposed in these lines (‘He who learns to write/Drives a horse and cart’) may well be seen as a banter of his financial meticulousness. The girl-child’s deprivation of higher education adds yet another ironic dimension to these lines. The Secret Adventures of Haridas is a sensational if not scandalous book written by Bhubonchandra Mukhopadhyay in 1903, representing life at Sonagachhi, one of the oldest red-light districts in Kolkata. Uma’s lines, referring to beautifully red flowers on slimy water, thus become an empathizing symbolic commentary on the condition of these public women.
This writing process climaxes when Uma scribbles on elder brother Gobindolal’s scientific article. Gobindolal’s writing here is the contrastive frame of male writing, strictly located within the rational sphere of discursive public writing. The narrator undercuts Gobindolal’s thinking by commenting that his article attacks the Western scientific precepts, not so much on the basis of logic, but merely by the means of a ‘romanchokar bhasha’ [‘exuberance of his language’ in Radice] (Ibid.). She writes in a big font on top of this piece—‘Gopal boro bhalo chhele, tahake ja deoa jay she tahai khay’ [‘So well-behaved is young Gopal/Whatever you give he eats it well,’ in Radice] (Ibid.). This line, alluding to Ishvarchandra Vidyasagar, is yet another of Uma’s ironic laughs directed at the male culture of writing. All these lines evoke an ideology of feminine inscription which subverts the male stress on scientific rationality with poetic interventionism. The apparent senselessness of Uma’s scribbles contests the male tradition of argumentative writing. This is not nonsense but non-sense. This writing maximizes the symbolic repertoires of the Symbolic order of language but this maximization of symbolism in this case delves into non-sense and obfuscates integral transmission. It is in this hyper-symbolic unreadability that this writing breaks with patriarchal reason.
Gobindolal trounces Uma for writing on his article and confiscates her pen as well. We should note here that it is none other than Gobindolal who eventually returns her exercise book to her, after being moved to repentance. So, the exercise book is re-circulated from within patriarchy; it is seen as the compromise of a consolation prize, inscribed within the patriarchal field. Uma’s poetic etchings continue as she appropriates the exercise book—‘Pakhi shab kare rab, rati pohailo’ [‘Birds are singing, Night is ending,’ in Radice’s rendering] (Ibid.). Her first expression of love is ‘Yashike ami khub bhalobashi’ [‘I love Yashi very much’] (Ibid.; p.164), written once again as an additament to the ‘Story of the Tiger and the Crane’ in Kathamala, copied again in her exercise book. This first expression, significantly, is not about some young boy, but their old maid. The word ‘love’ applied outside the family for the first time, refers to a woman and when the second proper name in relation to love appears in the ‘khata,’ it is ‘Hari.’ Tagore’s narrator promptly points out that it is not ‘Haricharan’ but ‘Haridashi,’ her school-friend. It is once again the name of a woman and not that of a man. After one year, when Uma is married off to Pyarimohon, the assistant-writer of Gobindolal, the note of patriarchal prohibition is clear in Gobindolal’s words—‘… Pyarimohoner kono lekhar upore khabardar kalam chalash ne’ [‘… And make sure you do not scrawl on any of Pyarimohon’s writings,’ in Radice] (Ibid.).
It is Yashi who takes her exercise book to her in-laws’ abode. The exercise book by this time has become an object of nostalgia—something that reminds Uma of her origin. It has become ‘pitamatar ankosthalir ekti shankhipto itihash’ [‘a brief record of parental affection,’ in Radice] for her (Ibid.). At her in-laws’ house the exercise book becomes the sole witness to her fundamental desire for returning to her mother. It becomes an intensely personal space outside social gaze where she can inscribe her purest desires in singular expressions. But finally her scribbling falls prey to the eyes of her sisters-in-law—Tilakmanjari, Kanakmanjari and Anangamanjari, names emerging straight from fairy tales. When this news reaches Pyarimohon, his reaction echoes the typical patriarchal dialectic between the aesthetic and public functions of writing a novel or a play and the domestic sphere wherein the supposedly real function of the woman lies. That the writing itself can inhabit a deeply domestic space in a gendered and existential sense, goes well beyond him. He mocks Uma by evoking the idea of a working woman going to the office with her pen. As against the simple non-theoretical bent of Uma’s writing, Tagore builds up an idiosyncratic theoretical register of male thought. He refers to Pyarimohon’s peculiar theory that if a woman becomes educated and starts to work, she virtually becomes a man and two men can never constitute a proper domestic structure. What is crucial about this theory is its pseudo-liberal and hegemonic nature. It pretends to argue against the defeminization of women and seems to maintain that an all-male domination is unhealthy for the domestic space. And yet this theory remains subtly regressive by continuing to limit women to a domestic space, from which writing, conceived as the public discourse of male prestige, is banished forever. Tagore’s women may not join an uprising against the sexist hierarchy of the private and public spaces qua the practice of writing, but they nevertheless appropriate that delimited space and make their presence felt through their signifiers.
The climax arrives when Uma is tremendously moved by the ‘agomonir gan’ or ‘the homecoming-song’ of Uma-Parvati, sung by a beggar-woman, one autumn morning. It is her act of copying this song in her ‘khata’ that spells doom for her. She has to part with her ‘khata’ and the text in it provokes great laughter in her sisters-in-law. In this last occurrence, Tagore uses the myth of Uma, one of the aspects of the great Indian Goddess, lord Shiva’s wife and someone who can be the gentle Parvati as well as the fiercely destructive Kali. It is the polymorphous nature of this mythical figure that contests the patriarchal strategy of feminine stereotyping. The mythical is used here as a counterpoint to the masculinist rhetoric of linear rationality. This mythification of Uma is rendered exclusively at the level of writing and the source of it is a typically feminine speech-act of the beggar-woman. The beggar-woman here can be seen as Julia’s Kristeva’s figure of the ‘abject’ (Kristeva, J. (1982) ; p.521), operating outside the domain of the masculinist social hierarchy of language. Pyarimohon’s response to his, reading the lyrics of the song from Uma’s exercise book, is not recorded in the story. This crucial omission in Tagore’s text tears an enigmatic hole in the established repository of masculine knowledge. Pyarimohon cannot unpack the truth in Uma’s circuit of desire. It hides beneath a garb of inconsequentiality, embedded in the mythic symbolism of the ‘agomonir gan.’ As Tagore’s narrator clarifies in the last sentence of the story, even Pyarimohon has an exercise book but it is completely different from Uma’s. It is the domain of his subtle theories of the world and there are no sensible persons around who would take it from him or destroy it.
If ‘Khata’ is interspersed with the poetic etchings of Uma, the entire text of ‘Streer Patra’ is a feminine discourse in an epistolary form. Though at the level of this form, it is a personal communication, Mrinalini’s discourse, unlike Uma’s, is an appropriation of the rational, argumentative and supposedly male discourse. Right at the outset, Mrinal clarifies that her need to write a letter to the husband is triggered by the ‘phank’ (Ibid.) or gap between the two of them. In the story, the epistle or the letter functions in this gap between the man and a woman, much like the man-woman relationship in Lacan, which can only be addressed through the edifice of the signifier in its localized structure, i.e. the letter. Mrinal is on a pilgrimage to Puri while her husband is back in Kolkata, doing his office work. She also states that this is not a letter from their ‘mejobou’ but from a woman who has an independent relation with the world and its creator. As she maintains, death will not take her away because it only takes away those who have some value and being a woman she is hardly considered valuable. The letter thus, is also a declaration of her deathlessness. The thesis of deathlessness is ironically juxtaposed with her death wish. Mrinal’s letter also declares suicide. It is her first, last, and only letter to her husband.
Mrinal’s beauty has its takers in the house of the in-laws but no real appreciation; they are most resentful towards her intellect. She secretly writes poems but nobody knows that she is a poet. Her poetic writings constitute a singular subjective space where her feminine identity flourishes outside the paradigms of patriarchy—‘… shekhane tomader andarmahaler panchil otheni. Sheikhanei amar mukti; sheikhanei ami ami’ [‘… at least there the boundary wall of the inner compound could not stop me. There lay my freedom, there I could be myself,’ in Prasenjit Gupta’s translation] (Ibid.; p.522). Mrinal appropriates the theoretical register of male discourse when she philosophizes about neglect. She observes that it is better for a woman to be neglected than to be cared for. Since the arrangement of patriarchy is always directed towards female suffering, the passing phases of care only increase the sorrows of suffering.
The real event, founding Mrinal’s subjectivity, is the arrival of Bindu [meaning a ‘tiny speck’ in Bengali], a fourteen-year-old girl, exploited by her brothers, in her maternal house. She is the sister of Mrinal’s eldest sister-in-law. Mrinal fosters this ordinary girl, who quickly becomes an eyesore for the whole family, including her own elder sister. Everyone is keen to humiliate her out of the house. Be it some red spots on her skin, immediately judged as chicken-pox or the allegation during the Swadeshi movement that she is a police-spy, everyone in the family, except Mrinal, is in constant attempt to banish or excuse her. Mrinal uses the expression ‘o je Bindu’ [‘After all, it was Bindu’] (Ibid.; p.525) repeatedly to indicate that her (wo)manhandling is premised on her singular identity as a gendered subject. The absolute singularity of her subjectivity is marked by the proper name ‘Bindu’ which hovers towards the common noun where the word signifies a tiny little speck in the horizons of sight. Patriarchy’s problem with a woman lies exactly where she is named in a singular way as the Real woman, where Bindu is the Real Bindu.
The Mrinal-Bindu relationship in ‘Streer Patra’ is an identificatory relation which is inscribed entirely on the feminine axis and I suggest it can be seen as a compensation for Mrinal’s failed motherhood. It is part of the entire field of woman-identified experiences Adrienne Rich calls the ‘lesbian continuum’ in her 1980 essay ‘Compulsory Heterosexuality and Lesbian Existence.’ This is how Rich defines the ‘lesbian continuum’:
I mean the term lesbian continuum to include a range–through each woman’s life and throughout history–of woman-identified experience; not simply the fact that a woman has had or consciously desired genital sexual experience with another woman (Rich, A. (2010); online).
In her letter, Mrinal gives an analogy of heterosexual love for their relation, at least from Bindu’s side—‘Bhalobashar erakam murti sangshare to konodin dekhini. Boiete porechhi bote, sheo meyepurusher modhye’ [‘I have never seen such an embodiment of love in real life; I’ve read of it in books, of this kind of intense attachment, and, there too, between men and women’] (Tagore, R. (2002); p.525). Marriage finally becomes the strategy of displacing Bindu within the household. Bindu does not want to leave Mrinal but Mrinal knows that in her absence there would be no one to take care of her. On the other hand, the exploitations that this ‘ugly’ girl may have to face in her in-laws’ house rankles her greatly.
Bindu gets married finally but it transpires that her husband is ‘mad’ and it is her mother-in-law who forced this marriage against the wish of her father-in-law, who himself has left on a pilgrimage to Kashi. Although the figure of the mad husband is left like a narrative tangent, the idea of madness signifies a social otherness analogous to the Othering of women in the patriarchal arrangement. After an eventful passage of time, when Mrinal plans an escape for Bindu with the help of her younger brother Sharat, he returns only to communicate the news of her suicide. As Sharat reports, Bindu has burnt herself to death—a so-called ‘fashionable’ way of dying. Incidentally, Bindu had also written a letter to Mrinal but it was destroyed by her in-laws. Mrinal’s letter, which forms the entire text of this short story is, as it were, a reinforcement of Bindu’s missing letter.
Towards the end of the story Mrinal contemplates death in general and her own death in particular. She sees it in the offing as she stands at the Puri Beach, looking out into the vast expanse of sea. This writing on death becomes a transgressive political strategy of feminine emancipation. Death in its universality far surpasses the social terrain of patriarchy. It is that which goes beyond all cultural specificities and gives a generic feminine identity to a woman—‘Mrityu tomader cheye boro…shekhane Bindu kebol khurtuto bhaier bon noy, kebol aparichito pagol shamir prabanchito stri noy. Shekhane she anonto’ [‘Death was stronger. In her death Bindu has become great; she’s not a mere Bengali girl anymore, no more just a female cousin of her father’s nephews, no longer only a lunatic stranger’s deceived wife. Now she is without limits, without end’] (Ibid.; p.530). The movement from Bindu, a tiny speck, to the vast occean has an expansive semantic trajectory present within it. Death is the point of intersection between the finite speck and the infinite space of eternity. Let us also note how death crystalizes the unlimited and endless enigma of a woman which is consistent with Lacan’s point in Seminar XX that there exists a distinct feminine jouissance (a masochistic enjoyment, combining pain with pleasure, which orders the movement of desire) which is ‘supplementary’ in relation to the masculine one. This death, in Mrinal’s eyes, is only a celebration of life—‘Oi to mrityur hate jiboner joypataka urchhe’ [‘… but the proud standard of life flies from the hand of death!’] (Ibid.); it is death that binds Mrinal and Bindu in the final run. The reference to Mirabai’s song further extends the ‘lesbian continuum’ of all female identification. As the song demonstrates, when the father, the mother, and everyone else departs, there is always one thing (Mira herself in this case) that persists. Mrinal is in the process of removing everything in her embracing of death and yet something still remains after it, like an enduring urn of ashes. The letter to the husband suffices this remainder when everything else is stripped away. That is what matters. The letter, thus, ends with a triumphant assertion of death which has become life—‘Amio banchbo. Ami Banchlum’ [‘I too will be saved. I am saved’] (Ibid.; p.531). The incommunicable experience of death which resists linguistic codification is on the side of the Lacanian Real while the epistolary bridge of communication in a written form of desire, reinforces the Symbolic. In this joint of the Real and the Symbolic, we encounter the Lacanian impossibility of communicating with the Other sex without the meaning effect of the signifier.
Death, importantly enough, is the point where all writing has to stop. There can only be a writing on death or a writing which prefaces death. There can never be a writing of death. Therefore, Mrinal’s meditation on death is also a meditation on the limits of her own writing or in other words, on the limits of what can be written insofar as death is the unwritable per excellence. It is this beyond of the writable, the speakable and the thinkable that stumps patriarchy. Insofar as Mrinal’s letter embodies her death drive, like the supplementary feminine jouissance, it can only be experienced and cannot be interpreted. Patriarchy, like all other forms of totalitarian ideology, is built on a fantasmatic notion of the whole and I would claim that the point where death (of a woman) interrupts and even haunts this discourse is where this fantasy of the whole collapses. The death-point is a discourse in lack and it is impossible to totalize this point. It is in this sense that Lacan’s tragic heroine Antigone, in all her ‘splendour’ remains a complete enigma to the phallic law of Creon’s patriarchal city state. What is at stake in Sophocles’ play as well as Tagore’s story is a transgressive feminine jouissance driving the movement of a (s)heroic tragic desire.
‘Nashtanirh’ develops the paradigm of writing both thematically and tropologically in its view of the man-woman relationship. The trio of Bhupati, his young wife, Charulata and brother-in-law Amal is engaged in a circuit of desire that corresponds with the circuit of textuality, woven around them. Tagore contrasts the private sphere of feminine writing with the public sphere of male writing and questions the inherent sexism of this binary opposition. At the outset, we are told that Bhupati’s decision to edit an English newspaper is only driven by his ‘shakh’ or his whim (Ibid.; p.347). He does not need to write. It is the rhetorical lure of a public discourse that he cannot let go. Tagore reflects that the bar that alienates Charu from Bhupati is a ‘kagajer abaran’ [‘paper cover’] (Ibid.; p.348), which is another way of saying that the discursive space of writing surface produces an interval between the man and a woman instead of conjoining the two. We can translate this into Lacan’s (in)famous claim in Seminar XX that there is no such thing as a sexual relation because the phallic signifier incarnates the unbridgable gap between the two sexes and introduces a third element in speech instead of building a rapport of the two. Bhupati is always busy with his newspaper and does not have time for his wife and hence it is the writable surface of a page that stands between them. Charu has a taste for reading and that is how she spends her solitary time. Amal becomes Charu’s sole companion in her hours of solitude and they become great friends. It is their plan to make a garden out of the waste land in the interior of the house that opens up the tropes of private writing. Charu insists that Amal should write a story about all the planned details of this prospective garden and that story, once written, will only make sense to them—‘… amra dujone chhara keu bujhte parto na, besh maja hoto’ [‘… no one apart from just the two of us would have understood it; it would have been great fun’ in my translation] (Ibid.; p.351). Amal had started to scribble little pieces in his ‘khata’ long before. He soon shows Charu one of his scribbles. Importantly, this first piece is addressed to the virgin surface of a blank page, awaiting inscription.
When this piece, dedicated to the the exercise book, is published in a popular Bengali magazine ‘Shororuho,’ the private space of writing between Charu and Amal is invaded by the public one. This displeases Charu and more importantly, Amal cannot understand this displeasure—‘Amaler lekha Amal ar Charu dujoner shampatti. Amal lekhak ar Charu pathak. Tahar goponatai tahar pradhan rash. Shei lekha shakale poribe ebong anekei proshongsha koribe, ihate Charuke je keno etota pirha ditechhilo taha she bhalo koriya bujhilo na’ [‘Amal’s writing is the joint property of Amal and Charu. Amal is the writer and Charu the reader. Its secrecy is its primary quality. That writing will be read by all and even appreciated by quite a few—this thought was making Charu suffer. And why did this cause suffering, Amal could not even understand properly,’ in my translation] (Ibid.; p.353). Fan-mail begins to arriving and Charu soon loses her role as the singular motivator and reader of Amal’s writing.
Manda, Charu’s sister-in-law, is Amal’s other reader in the house and the more Amal wants to read his writings to Manda, the more it irritates Charu. It is to distinguish herself from Manda that she starts writing. Initially the literary spectre of Amal looms large on her and her piece on rain-clouds turns out to be an echo of Amal’s piece on the moon. She decides to change her subject to get rid of Amal’s influence and writes a piece called ‘Kalitala,’ drawing on her childhood memories of a place long lost in time. From the moon to the dark and mysterious temple of the goddess of destruction, Kali, the shift of subject suggests subversion. Amal reads the piece and decides to send it to a magazine for publication. After a lot of hesitation, Charu yields but soon tables the proposal of a hand-written magazine where only she and Amal would write. Her condition is that Amal cannot write in any other magazine. Charu proposes to come up with only two copies of every issue—one for each. This is yet another of Charu’s formulations of a closed and personal textual circuit where the two of text overlaps with the two of love. Amal does agree but deep within, he has lost his interest in secrecy a long time back. We encounter a dialectical tension here between a feminine desire for a private world of shared textual spaces and the masculine desire for a rhetorical accomplishment of writing in the public sphere where the wider field of the Other beyond the amorous couple is addressed and activated.
When Amal secretly sends Charu’s pieces, preserved for their secret hand-written magazine, to ‘Shororuho’ and they get published there, she gets upset with Amal. It is like the freedom of one’s beloved pet birds. It brings both liberty and pain with it. Not only does Charu’s writing come out in ‘Shororuho,’ in another magazine, ‘Bishwobandhu,’ a critical review of contemporary Bengali writing genuinely appreciates Charulata’s style of writing. The critic slams Manmatha Dutta and Amal’s artificial and ideational style and praises Charu for the unabashed simplicity, pastoral melody and imagistic qualities of her writing. This review once again draws our attention to the contrast between discursive male writing and the non-discursive and affective essence of the feminine inscription. But this appreciation accompanying flack for Amal does not please him and to make Charu jealous, he starts reading his pieces to Manda once again.
In the meantime it is Bhupati who wants to give attention to his neglected wife by listening to her writing. The entry into the circuit of desire once again necessitates an entry into the textual circuit. However, Charu does not yield on the first occasion. On the other hand, Charu returns to Amal’s style of writing to impress him afresh—‘Charu etuku bujhiyachhe je, tahar shadhin dhancher lekha Amal pachhanda kare na’ [‘Charu has understood this much that Amal does not like her own independent way of writing,’ in my translation] (Ibid.; p.368). When Bhupati is hurt by the betrayal of his dear Umapada and wants Charu to nurse his wounds, she is busy writing her piece to reconcile with Amal. She hides her writing once again from Bhupati. He is left alone, because he is outside the circuit of textuality. Amal however realizes Bhupati’s loneliness and feels guilty of attracting all of Charu’s attention. At this point in the story, Tagore uses the image of a deep dark hole, into which a traveller is just about to fall. But he is saved as the fog lifts at the very last moment and he encounters the chasm. This chasm is read on the surface of Bhupati’s face or if I may say, it is misread on Bhupati’s face. Throughout the story, Tagore uses the face as a textual metaphor where lines are read, left unread or misread. Amal reads the lines right but gets the source wrong. He withdraws himself from Charu.
Shortly thereafter Bhupati finds a bride for Amal with the prospects of a foreign settlement if the marriage happens. Amal readily accepts the proposal because it gives him the opportunity for a much-needed distance from Charu. Amal keeps ignoring her before departing and tells her to take care of Bhupati in tough times. As Amal goes abroad, Bhupati’s life also changes after doing away with his newspaper due to finance problems. He regrets having not taken enough care of his wife; as a result of which now it seems so difficult for him to resume a dialogue with her. After quite a few futile attempts, he chooses an entry point through the textual circuit. Bhupati starts reading Tennyson, Byron and Bankimchandra Chattopadhyay’s works and tries to impress his wife by reading from Tennyson, but it falls woefully wide of the mark. Charulata seeks refuge in the deepest and darkest chasm of her mind with an acute pain of separation from Amal, at the same time, wondering at the intensity of the pain.
As Charu slowly gets over the crisis of Amal’s lack and starts to take care of her husband, Bhupati wants Charu to start writing again. He himself starts to write in order to inspire his wife. He writes something in imitation of Charu and says it is the writing of a friend. Charu understands the lie but does not express herself. That her husband has to resort to the textual to get her attention makes her feel uncomfortable. She still praises the piece and the inspiration from his wife soon translates into pages of writing in Bhupati’s exercise book. It is another textual circuit, this time an epistolary one, that disturbs the stabilizing domestic space of Bhupati and Charu. Amal writes his first letter from overseas but hardly mentions Charu in it. Charu gets tremendously upset over this neglectful omission. Time passes and Amal tells them that his exam is on the cards and he would not able to write for a while. Charu becomes anxious as Amal’s last letter slowly becomes an event of remote past. She secretly sends Amal an expensive telegram abroad by mortgaging her jewellery. Amal’s reply reaches Bhupati’s hands and a deep doubt sets in. Bhupati burns his writings in rage and disgust at the betrayal of both public and private worlds. This may be seen to mark the failure of male writing when it tries to operate inside the private sphere. Not only does Bhupati throw his own writings into the kitchen fire, but he also throws Charu’s writings into it.
From this point onwards, Bhupati becomes extremely anxious about the crevices of Charu’s desire. There is an opening into a different order of desire but Bhupati cannot find the key to that order. The anxiety passes on to an interpretive desire and he considers Charu’s maneuvers false and deceptive. This ascription of falsehood and deception on that which is beyond understanding is Bhupati’s sad case of patriarchal forcing. He feels sorry for Charu’s laboured deception and her suffering in trying to put on the mask. The patriarchal assignation of feminine masquerade is turned on its head and now it is the mask which gives pain to Bhupati. But what he considers Charu’s mask is a different order of authenticity, the Real of her ‘face’ which remains inaccessible to him. There is genuine appreciation of Charu’s stolidity in Bhupati but if not consciously, in the unconscious at least, it all goes back to the ascription of a victim-like status of the woman.
Bhupati gets the job as an editor of a new newspaper in Moishur and decides to leave without Charulata. As Bhupati is on the verge of leaving, Charu asks him to take her along—‘Amake shange niye jao. Amake ekhane phele jeo na’ [‘Take me along with you. Do not leave me here,’ in my translation] (Ibid; p.383). The parting hands and the distancing posture say it all. He misreads Charu’s desire to be a desire to leave a mansion that reminds her too much of Amal. Bhupati replies that he cannot take Charu with him. It is the clarion call of an other textual circuit (this time, a public one fortunately) at Moishur that awaits him. But this time it is a return to the loveless textuality of Bhupati’s earlier newspaper. Tagore’s narrator uses the last crucial textual metaphor here while commenting on Charu’s facial expression in reaction to Bhupati’s ‘no’ or his denial to take her with him. The narrator compares her face with a dry and blank page—‘Muhurter modhye shomosto rakta namiya giya Charur mukh kagajer moto shushko shada hoiya gelo,…’ [‘In a moment all the blood trickled down and Charu’s face became like a dry and pale paper,’ in my translation] (Ibid.). The metaphor rounds off the textual motif of ‘Nashtanirh’ as it returns to Amal’s initial image of the white virgin page, depicted in his first piece. Bhupati is stricken by the opaque quality of Charu’s face, which makes him apprehensive about his reading. He reverses his decision and says he would take Charu to Moishur. But this time, it is Charu who says ‘no’ and the story comes to an end.
Charu’s ‘no’ of resistance is the culmination of her discourse in ‘Nashtanirh.’ Her face is a surface where little ciphers of desire appear, disappear and reappear against which the patriarchal game of misreading continues. The final similitude between her face and the blank page establishes a post-textual void. This is not a virgin page. It is a page on which letters have been erased. After the erasure, there remains a spectral trace of ciphers on this page and it is this enigmatic trace of feminine desire that remains a closed self-sufficient circuit, unreadable to patriarchal hermeneutics. Bhupati can never understand the real significance of this final ‘no.’ Tagore’s decision to end the story at this point, thereby excluding Bhupati’s reaction (much like the exclusion of Pyarimohon’s reaction at the end of ‘Khata), reinforces the point.
In Seminar XX Lacan defines the supplementary nature of feminine jouissance thus:
There is a jouissance that is hers (à elle), that belongs to the ‘she’ (elle) that does not exist and doesn’t signify anything. There is a jouissance that is hers about which she herself perhaps knows nothing if not that she experiences it – that much she knows. She knows it, of course, when it comes (arrive).
The fact that this jouissance can be experienced and can not be explained recalls Charulata’s opaque face, Uma’s senseless etchings and even the appropriated male discourse of Mrinal which is hauntingly punctuated by the Real of death. As we have seen above in these three short stories, Tagore mobilizes the writing of women to interrogate the private-public binary, prevalent in the male-dominated practices of writing. I have argued that his female figures, motivated by a transgressive jouissance of supplementation, subvert man’s phallic jouissance. The order of the indecipherable is crucial in this context and Tagore’s women remain faithful to the Lacanian Real of the Other jouissance by re-marking their unreadability vis-à-vis the patriarchal drive for phallic interpretation.
Santosh Chakrabarti, Studies in Tagore: Critical Essays (New Delhi: Atlantic, 2004)
Ranjana Ash, ‘Introducing Tagore in Multicultural Education in Britain’ in Rabindranath Tagore: Perspectives in Time (London: The Macmillan Press, 1989)
Supriya Chaudhuri, ‘A Sentimental Education: Love and Marriage in The Home and The World’ in Rabindranath Tagore’s ‘The Home and The World’: A Critical Companion (New Delhi: Permanent Black, 2003)
Rabindranath Tagore, Galpoguchho: Akhanda (Kolkata: Shatyanarayan Prakashani, 2002)
Jacques Lacan, Book XX: Encore 1972-73; On Feminine Sexuality: The Limits of Love and Knowledge (London: Norton, 1998)
Julia Kristeva, Powers of Horror: An Essay in Abjection (New York: Columbia University Press, 1982)
Adrienne Rich, ‘Compulsive Heterosexuality and Lesbian Existence’
http://www.terry.uga.edu/~dawndba/4500compulsoryhet.htm (accessed July 25, 2010)
Lacan, Book XX, 74.
NOTE – this is a re-print:
Shifting Identities: Constructions and Re-constructions of the Feminine in Indian Literatures, ed. Prof. Sutapa Chaudhuri, Books Way: Kolkata, 2011.
University of East Anglia
Spring 2007: the high-water mark of self-confidence for economic neo-liberalism. In March, both Henry Paulson and Ben Bernanke publicly stated that they saw no danger of recession, and that the subprime fiasco had been ‘contained’. As late as mid-May, with the sub-prime crisis in full throe, still Bernanke felt able to say this: ‘Importantly, we see no serious broader spillover to banks or thrift institutions from the problems in the subprime market.’ In July, Paulson claimed: ‘This is far and away the strongest global economy I’ve seen in my business lifetime’; and on August 1st, ‘I see the underlying economy as being very healthy.’ Neo-liberalism remained a movement triumphal around the world. No bunch of poverty-stricken mortgage-defaulters – who could conveniently be blamed for the little local difficulty – were going to derail this ideology.
Sure, occasionally the masters of the universe would admit that all was not entirely rosy: thus for example, again in 2007, Nick Stern announced that manmade climate change is the biggest market failure in history. Then again, some see manmade climate change as the greatest market success in history – for what are markets for, if not to externalise costs as much as humanly possible, in order to stay as competitive as possible? One person’s externality is another person’s (or rather: firm’s) profit-margin.
Spring 2007: Bryan Caplan’s ‘The myth of the rational voter: Why democracies choose bad policies’ is published, to vast acclaim. According to the New York Times it’s ‘The best political book this year.’ Barron’s Magazine for instance finds it ‘Scintillating, outstanding.’ Gregory Mankiw of Harvard University, former chairman of the President’s Council of Economic Advisers and author of the best-selling economics textbook ever, remarks that ‘Caplan offers readers a delightful mixture of economics, political science, psychology, philosophy, and history to resolve a puzzle that, at one time or another, has intrigued every student of public policy.’ And so on (and on).
So, what is this idea that is proving so popular? What is ‘The myth of the rational voter’? It should be noted immediately that the title of Caplan’s book is somewhat misleading. Caplan believe that voters are reasonably rational in one very important sense – in the sense of self-interested instrumental rationality, the sense central to the ‘Public Choice Theory’ school of thought, of which Caplan is now a very prominent (if iconoclastic) member. Core to his case, in this book, are well-known truths of political science in an age of consumerist individualism, such as that voters are self-interestedly rational not to spend too much time trying to decide what the most rational way to vote is (because they are very unlikely decisively to influence the outcome). Voters are ‘rationally ignorant’.
To justify his title, Caplan clearly needs to go further than this. He first does so by suggesting that it is a myth that voters try their best coldly to work out what is rational for everyone and then vote accordingly.
So far, so obvious. Indeed, if this were merely meant to mean that voters vote in part according to their emotions, their passions, their general way of seeing and feeling the world, then it should certainly be admitted: though it is still not entirely clear that such an admission should really be taken to show that voters are not rational. For instance: according to the arguments made by Drew Westen in his hugely-influential ‘The Political Brain’ (Obama himself has read it, so we are told), if we mean by ‘rational’ what the word really means, and not a rationalistic caricature thereof, then voters voting with passion generally are voting rationally. For voters generally vote their political values, which are what they want to see society uphold. What could make more sense – what could be more truly rational – than that?
Caplan allows that voters decisions are partly ‘expressive’ / ‘symbolic’ (pp.137-9), but he doesn’t seem much interested in values or passion. What Caplan actually has in mind in speaking of pervasive voter irrationality – what he most consequentially adds to the ‘rational ignorance’ hypothesis – can be seen most clearly in a central chapter in the book, entitled ‘Evidence from the survey of Americans and economists on the economy’ (Interesting to note that economists are apparently exempt from belonging to a nation). Caplan’s method in this chapter is basically as follows: He contrasts the views of the public with the views of the enlightened (i.e. mainstream economists), and suggests that this shows that the public are irrational. In the particular sense that (according to Caplan) the public tend to have false beliefs about the economy and about other crucial matters of state.
This would perhaps be a convincing method if one were convinced by what economists believe to be true. Those beliefs include that top executives are not over-paid (p.64) and also (Caplan claims) that the minimum wage is inefficient and economically harmful (p.96). In a Caplanian economically-‘rational’ world, evidently, income differentials – and bankers’ bonuses – would be even more obscene than they are today. Moreover, Caplan’s central example, with which he opens the book (‘Protectionism is a classic example. Economists across the political spectrum have pointed out its folly for centuries, but almost every democracy restricts imports’; p.1) and which he dwells on over and over including at length in this key chapter, inspires no confidence, at least not in my part of ‘the political spectrum’. ‘Free trade’ is a nostrum beloved of neoliberal economists; but it was not practised by most powerful economies on their rise, is rightly feared by countries concerned about the take-over of their economy by other more powerful economies, and is part of the problem rather than of the solution in relation to vast ‘market-failures’ such as dangerous climate change. Caplan trots out ‘the law of comparative advantage’ (p.38, p.66), but omits to consider whether it still applies in a globalised world of free movement of capital, or whether such a world (our world) is instead a world in which ‘absolute advantage’ is all that counts. For under-performing countries cannot count on retaining their capital at all in order to do even what they are second-best at. (In fact, Caplan proposes an alternative ‘explanation’ for why poor countries stay poor – that their voters are, allegedly, even less ‘economically literate’ than rich countries’ voters (p.158).) Protecting the local, protecting labour, protecting our environment, protecting economies from foreign dominance, protecting the global economy from an excess of inter-dependence and a lack of resilience – in a world subject to the dominance of globalised capital, very many people plausibly argue that these protections are good things, not bugbears. The jury is still out, in the case of Protectionism vs. Free Trade. But Caplan disagrees: it is for him a strictly ‘empirical’ finding (p.147) that ‘voters underestimate the benefit of foreign trade’ across the board. It tells one a lot about Caplan’s standpoint and method that he is really not interested in the jury’s verdict, if it disagrees with that of the judges (i.e. of economists). The jury metaphor is in fact Caplan’s, not mine. He writes that ‘the jury is in’ (p.10) and that it has ruled that ordinary people do not understand the unrivalled benefits offered them by unregulated privatised markets, free trade, etc. . But he seems in the process to have forgotten what a jury is. It is the verdict of one’s citizen-peers – not of one’s own privileged workmates or one’s friendly ‘experts’.
If Caplan didn’t believe in markets so much, then one surmises that he would turn to the next best thing, as an alternative to democracy: rule by economist-kings. For Caplan, the basic test of rationality – which the people fail – is the following: they don’t agree with neo-liberal economists about political economy.
In order to demonstrate pervasive voter irrationality in the sense of voters having pervasively false beliefs, Caplan would need to actually show that those beliefs were false. But he simply doesn’t do this; he just presupposes that mainstream economists are nearly always right. Caplan is quite correct that the truth often hurts, and that people often tend to avoid it – but he fails to begin to prove that this truth is truer of ordinary voters than it is of neoclassical economists.
To support the thought that voters don’t have any ‘rational’ motive to become politically-literate, and thus tend rather to use their vote to indulge their (alleged) prejudices – that they are ‘rationally irrational’ -, Caplan impresses upon us how unlikely we are to influence the outcome of elections. (He patently overplays his hand in this regard at more than one point, as when he claims that the price of political irrationality is ‘zero’, and that voting has ‘no practical efficacy’ (p.132, emphasis added). There is a crucial and obvious difference between low probability and no probability. Ordinary people (voters) understand the difference: see below.)
In the literature, this is known as the ‘rational voter paradox’: it is allegedly so unlikely that one will swing an election that it is irrational to vote at all. But perhaps professional political scientists should get out more: If they did so, into the real world of elections, they would find some telling counter-examples to this alleged paradox…
I was closely involved in one myself, in the local elections of Spring 2007. In Norwich, England, where I live, the Green Party failed to gain a seat from the Liberal Democrats by one vote. That made the difference between the Greens (who had 10 seats) becoming the official Opposition on the Council, and the LibDems (who had 11) remaining the Opposition. Norwich being one of very few strongholds of Green support in the country at that time, this narrow-as-can-be result made national news. For a little while at least, it held back the Green advance in Norwich, and probably nationally too.
Note this: if, as in this case, an election gets decided by one vote, then every single voter (and non-voter) decisively influenced the outcome. Anyone in the entire ward or constituency in question could have swung – did swing – that election.
But what about larger-scale elections? Caplan says this (p.131): ‘[I]n elections with millions of voters, the probability that your erroneous policy beliefs cause unwanted policies is approximately zero. The infamous Florida recounts of 2000 do not undermine this analysis. Losing by a few hundred votes is a far cry from losing by one vote.’ Well, not that far a cry. For Bush of course lost the election on every single measure except the one that the Supreme Court chose to stick with. There must surely have been a point, short of coming down to the very last vote, at which it would have been impossible for them to have sustained the fiction that Bush won. Maybe we were very close indeed to that point. Possibly if as few as 10 or 20 people less had voted Republican in Florida, and had swung Democrat instead, then the Supreme Court would have found it impossible to hold their thin red line against Gore.
While indeed it isn’t true to say that any single individual swung that election – that entire national Presidential election in 2000 – by their vote alone, there will surely be extended families that did so, or certainly small villages, or single activists in terms of the votes they swung. Think it through for a minute: an election which decisively influenced the course of the most powerful nation on Earth, an election that (for instance) decided life or death for hundreds of thousands of Iraqis, and which helped decide the very shape of the entire world’s economy in the noughties, actually was decided by every single township in Florida. If you were a Floridian on that day, then your vote together with that of your workmates or your neighbours literally made history.
So: it’s true that one’s chances of being the decisive influence upon an election outcome are very small. But it’s also true that it happens (and that when it happens as it did in my city, in Spring 2007, every last individual was crucial to the outcome). And that, when it happens, the effect can be very large. IF your vote makes the difference, then you can make quite probably the biggest difference anything that you ever did in your whole life makes.
So it is extremely unclear whether the ‘rational voter paradox’ really is a paradox, at all. Perhaps those pesky voters aren’t so irrational in bothering to turn out…
As it happens, if Caplan’s policy recommendations were acted on then one result would be that there would be more elections decided by just one vote than there already are. For Caplan is keen to keep the number of people voting down, on the grounds that those least likely to vote in elections are also on average the least literate and the least likely to agree with economists. On these grounds, he bluntly opposes ‘get-out-the-vote’ initiatives (p.157). He also notes (p.154) that men are more likely to agree with economists than women. Disappointingly, on this occasion he doesn’t draw the ‘logical’ conclusion – he fails to recommend the abolition of women’s suffrage…
Of course, the ‘rational voter paradox’ would remain a paradox after all, if the difference between one candidate winning and another doing so were of little or no consequence. And of course the alarming consequence of Market-Driven Politics is that it tends to have just that effect. When both candidates that one is choosing between are card-carrying members of the American Neo-Liberal Union, then why bother voting? Caplan’s attempt to entrench neoliberal economics as each nation’s basic wisdom and to allot elected politicians even less power than they already have is calculated precisely to reduce the voter’s incentive to bother educating themselves about politics, to vote rationally, and so forth. It appears that Caplanian thinking (and ‘Rational-Choice’ thinking in general) may be, as Karl Kraus might have put it, the very disease of which it takes itself to be the cure…
Spring 2007: the high plateau before the storm. If a week is a long time in politics, then 100 weeks plus must be some kind of eternity. For so much has changed since then. Everything from Gordon Brown being forced to ‘save the world’ by nationalising the banks, and the demise of all the independent U.S. investment banks and all the demutualised former building societies in this country, to the end of Iceland’s dream, and the beginning of Greece’s nightmare, and Cameron and Clegg’s ‘savage cuts’ here in Britain, and perhaps more imminent sovereign debt and commercial debt crises – and on we go.
When Caplan’s book first appeared on the shelves, his thesis of the superiority of markets to democracy must have seemed so natural, to many. For what is the essence of neo-liberalism? Perhaps this: an endless presumption in favour of markets.
In the final chapter of his book, Caplan attempts to rebut the charge that he rightly anticipates his readers are most likely to make against him: that of market fundamentalism. But his rebuttal isn’t likely to convince, given his willingness still to say things like this: ‘In the 1970s, the Chicago school became notorious for its ‘markets good, governments bad’ outlook. One could interpret my work as an attempt to revive that tradition. …Placed on a foundation of rational irrationality, perhaps the Chicago research program that Friedman inspired can live again.’ (p.197). It is rather unclear what would count as ‘market fundamentalism’, if a revived full-blown ‘markets good, governments bad’ Chicagoism doesn’t qualify.
Caplan says that his fundamental presumption – markets good, democracy bad – can in theory be overcome; but the burden of proof is heavy in the extreme. Moreover; writing as I am doing in mid-2010, surely one can’t any longer presume in favour of markets. In fact, looking at Obama’s America for instance, it rather looks as if big government, and even democracy, is back. And in that context, a book like Caplan’s that quotes Ayn Rand and Ludwig von Mises as authorities deserving of uncritical approval, while treating Islamic banking as a joke (p.33) and wheeling out Marxian exploitation theory in passing as an example of an ‘obviously false’ theory (p.118), runs the clear risk of already being past its sell-by. It is perhaps worth adding that the new Preface to the Paperback edition of this book, which appeared in August 2008, fully a year after the financial crisis ‘debtonated’, dwells a great deal on the favourable reviews that the book had already received, and also on the main lines of objection that had to date been offered to it, but contains not a word on whether the events of that past year might have cast some little doubt on economists’ faith in markets. It certainly doesn’t contain words which might seem appropriate in relation to that faith, words such as ‘hubris’.
Returning then to the titular claim of Caplan’s book: that voters don’t vote in a way that will be rational for the collective. Here is how he puts it near the start of the book: ‘In my view, democracy fails because it does what voters want. In economic jargon, democracy has a built-in externality.’ (p.3) As Caplan sums up the argument of his entire book in his Conclusion (p.206), ‘Democracy is a commons, not a market’. It is a truth universally acknowledged by neoliberal economists that any commons is in want of a market, or of being substituted for by a market. …But again, what a difference a couple of years makes.
In October of 2009, Elinor Ostrom won the so-called Economics Nobel Prize for her important work…thoroughly debunking the tragedy of the commons myth. This is by no means the first such thorough debunking – in fact, Polanyi (in ‘The Great Transformation’) did a fine job at debunking ‘the tragedy of the commons’ well before Garrett Hardin even invented it. But it must be galling for neoclassicals to see their own beloved quasi-Nobel won by explicitly commons-tragedy-debunking work.
Caplan is quite right to see that politics/democracy is not a market, and in that respect he decisively advances upon the economic-theorists-of-democracy. But we can now see that the conclusion that he draws from this observation, the conclusion that democracy should be largely substituted for by markets, is simply unwarranted, as well as being repugnant and potentially-disastrous. Let us see in more detail why.
Under what circumstances does a commons work well, and not suffer from the ‘tragedy’ that Hardin taught us to fear and expect and that Caplan expands upon? Polanyi, Ostrom et al tell us that the answer is: when people have established practices of working complementarily or together that either accidentally or (best of all) deliberately foster the common good. Democracy is supposed to mean: rule by the people. So: we need to build a people, not just a sea of individuals. The problem then with Caplan’s book is that it recommends moving in precisely the opposite direction. It sees democracy as a hopelessly-flawed attempt at summing individuals’ preferences, and proposes instead an allegedly better vehicle for summing individual rationalities: (more) markets. It would make us more thoroughgoingly consumers than we already are, and strip out the remaining elements of civic identity that ordinary people have.
How can one build a people, a demos? It’s not going to happen overnight; but it is already happening in some places. One powerful example is participatory budgeting, an experiment begun in Porto Alegre in Brazil, which is gradually spreading around the world (I know, because we are just now starting it on a small scale in Norwich). In this version of ‘deliberative democracy’, politicians gradually grant their constituents the right to determine how a progressively larger portion of their budget is spent. To the surprise, presumably, of the Caplans of this world: it works. Other versions of deliberative democracy such as citizens’ juries are also frequently proving successful.
So, rather than conclude from the people’s alleged foibles and failings that we should favour markets over democracy, perhaps we ought to actually try bringing about democracy (the rule of the people) instead. The point could be put thus: perhaps ‘democracies’ would be less likely to choose bad policies if they actually were democracies.
And here, one very hopeful point that emerges from Caplan’s book is that voters are generally altruistic (p.151). The ‘self-interested voter hypothesis’ is largely false; economists and public-choice-theorists have wrongly assumed that voters don’t vote for what they see as the good of the people. Caplan fears that this makes things even worse – for systematically misinformed altruistic voters will impose a worse societal outcome even than misinformed selfish voters (whose false beliefs will tend to cancel each other out more). But this fear can be turned around: to the extent that voters are not systematically misinformed (and I have argued that Caplan grossly over-estimates the extent to which they are, at least when compared to mainstream economists), and to the extent to which such misinformation as there is can be overcome (and that ought to be our task, in building better educational institutions, alternative media, etc), we have every reason to believe that democracy could not only exist, but could work very well indeed.
A key obstacle that needs overcoming if there is to be demo-cracy is the systematic misinformational bias of the media in favour of the interests of their main customers. Namely: of their advertisers. But while Caplan sees media-bias as a factor in the failure of actually-existing democracy, he not only sees it as a relatively small factor, but misidentifies its nature. The main thing he actually says, in this connection, is: ‘Journalism is a business. If consumers prefer news that fits their prejudices, journalists have an incentive to cater to them.’ (p.179) True – but it is a worrying mark of economic illiteracy not to be aware that the key product of commercial media is: audiences. Audiences – that is, vast and segmented groups of consumers – who can be sold to advertisers. And what most advertisers care about is reaching audiences with plenty of purchasing power. (They also quite like to avoid those audiences being alerted to any faults in the behaviour of the firms being advertised; that is, of themselves.) Thus the commercial media systematically distorts politics and economics to maintain the interest – and to favour the interests – of its relatively rich readers. Whereas it doesn’t much matter to them whether poorer readers exist or not. Commercial imperatives in the business that is media are such that the consumer-preferences of the poor are of little significance. In not seeing any of this, Caplan fails badly at his own test of empirical economics adequacy. One suspects that, ironically, he fails for the very reason that he promotes in his own pages. He fails to think clearly about the economics of the mass media, which reveals the truth of corporate power and its inegalitarian and harmful effects on citizens and on the democratic polity. Thinking clearly about that economics and politics would strike against his (Caplan’s) own political prejudices.
And yet, and yet: the rave reviews of Caplan continue. His star just won’t stop rising. Late last year (Dec. 8 2009), for instance, an article appeared prominently in the Guardian newspaper entitled: ‘How voters’ whims could scupper Copenhagen: ‘Rationally irrational’ voters could stall any deal on the environment’. It was basically a puff piece for Caplan. To be fair to Caplan’s thesis, it is of course perfectly true that democracy has a dire difficulty dealing with long-term issues such as management of our climate. But then again, so do markets. The problem is fundamentally one of a short-termism which is just humanly very difficult to escape. Future generations don’t vote – but neither do they play the markets. There is a real issue as to whether publics in so-called ‘mature’ democracies can/could ever be got to accept the kind of policies that are needed now to keep the planet liveable. Getting people to really care about the future after they are gone is hard; but there is absolutely no sign that Caplan’s preferred alternative will change that state of affairs. In fact, all indications are to the contrary. Carbon-trading will not keep the coal in the ground the burning of which would seal the fate of the Maldives, and Tuvalu, and, further down the line, of all our children.
So one has to ask: why, apart from the enduring attractions of feeling cleverer than the mass of dumb voters, is Caplan becoming so beloved of his public? The answer, presumably, is that his readers, existing in and in a good number of cases profiting from a world still awash with neoliberal ideology, mostly still think that untrammelled capitalism is the only game in town. They have not realised that this vast market-failure that is the financial-cum-economic crisis changes the rules of the game.
Similarly with the climate crisis: tragically, all the solutions on offer at Copenhagen that had any ‘legs’ with the developed world and with the new China-India axis were based around carbon trading, just as Kyoto has been – and just as Caplan recommends (p.33). Sub-prime carbon, anyone? The world’s financial markets are desperately trying to go on living as if it were still Spring 2007… Whereas: If the lessons of the financial meltdown had really been learnt, then we wouldn’t be rushing to sell off banks that are still TBTF.
But the lessons apparently haven’t been learnt. And so it’s worth thinking through for a minute what the Caplanian prescription – of less democracy, and more markets -, would really mean. It would mean that commercial imperatives prevailed virtually everywhere. Thus it would mean, for instance, faster progress than countries like Britain have already seen toward the dismantling of a public health service. The NHS was set up with the votes of soldiers, labourers, and so forth. When the voters that count are only pounds or dollars, then full privatisation is on the way. For everything. (Indeed, Caplan goes so far to insist that even ‘property rights and contracts are possible without government approval.’ (p.194) His example for this, apparently offered without irony, is rather telling. It is that ‘That is why one drug dealer can meaningfully tell another, ‘You stole my crack’ or ‘We had a deal’.’ More truly laissez-faire – i.e. lawless – economics is apparently what we can look forward to, in Bryan Caplan’s brave new world.)
Caplan’s Minervan owl surely flew out of sight already, in 2008 at the latest. The myth of the rational market is dead as a derivative. It looks like, if we are going to have a future worth having, then we have no alternative but to find a way of restoring and building this worst of all possible systems except for all the other systems: democracy. The longer-term background against which it is probably most useful to see Caplan is as the latest incarnation of an idea that is endlessly popular with elites: the idea that the biggest danger for a ‘democracy’ is the people. It goes back to the debates in France over how far the Revolution ought to extend political rights, and in England and the USA about what it means to extend the suffrage to people who are ‘uneducated’ and presumptively not capable of putting the general interest before their own. Famous previous versions of this idea have been presented by Samuel Huntington, the Trilateral Commission, Leo Strauss, Joseph Schumpeter, Walter Lippman, and earlier by Edmund Burke and by some of the Girondins and indeed by most of the founders of the United States of America. These thinkers have argued that ‘democracies’ work to the extent that the great unwashed do not act on the rights they have in a democracy, such as voting, free speech, protesting, and so forth, and to the extent that an elite rules instead. The twists that Caplan adds are primarily these two: first, that the uneducated don’t vote selfishly, but the reason that they are not capable of putting the general interest forward effectively is simply that they (allegedly) have systematically wrong ideas about what is in ‘the general interest’ of the country; and second, that the best way to stop them from messing everything up is to deprive their votes of effectiveness by suborning them to as unfettered a market as is feasible. It is because of this second point that Caplan is a true neo-liberal ideologue, a very model of the kind of writer who bestrode the intellectual universe – in Spring 2007.
NOTE – this is a reprint:
A Critical Notice of THE MYTH OF THE RATIONAL VOTER:
WHY DEMOCRACIES CHOOSE BAD POLICIES
By Bryan Caplan
[Princeton, 276pp., £12.50, 2007 (new paperback edition, 2008), 978 0 691 13873 2]
– Review Essay of Caplan, The European Review
The opposing concepts of the Dionysian (hereon DI) and Apollonian (hereon AP) are central themes within Nietzsche’s first major work, The Birth of Tragedy (hereon BT). His contemplation of such opposing forces of nature are primarily used to analyse Greek culture in general, and Greek art in particular, stating that its role in Greek attic tragedy places these plays at Greece’s cultural pinnacle. However, his ideas invite further social and political contemplation as to their resounding implications. In this article I intend to outline Nietzsche’s conception of DI and AP, expounding on the role he assigns them in Greek tragedy, before approaching the necessary implications of these ideas critically.
What does Nietzsche mean by DI and AP? The latter is derived from the concept of Apollo, the Greek god of light, who is often said to rule over the realm of the self-conscious, and is thus strongly related to the idea of individuation, through which he provides the world around us with a sensible structure. In contrast we have Dionysus, god of festivals (among other things), ‘centred in extravagant sexual licentiousness’ where ‘the most savage natural instincts were unleashed’ (Nietzsche, 1993, p.147). Crucial to this imagining of Dionysus, this ’emerges as an expression of the feeling of ecstasy that accompanies the sense of loss of the individual self’ (Sedgewick, 2009, p.60). Quite contrary to the individuated sense of the AP, then. This sense of individuation, Nietzsche claims, takes the individual out of nature, away from the community of beings in which they reside, where the latter concept negates such an alienation by placing one firmly in nature, relishing natural instincts and in turn breaking down any and all social barriers, allowing the sense of community spirit to thrive naturally once more.
Nietzsche’s conceptions of the DI and AP are used at length in his discussion of art, with a particular focus of Greek tragedy. A key component to these plays (e.g. Oedipus) is that of the Greek chorus, a collection of performers who narrate or pass comment on the action in unison. It is this chorus that Nietzsche deems fundamentally DI. We can of course see how a unified performance represents his idea of deindividuation associated with the DI, but the connection goes further: the characters of the chorus ‘remain eternally the same,’ regardless of what may come to pass within the progression of the story (such as the passage of time itself). It is of Nietzsche’s belief that such a ‘fictitious natural state’ will cause the audience to feel ‘nullified’ and create a similar feeling of having their individuality stripped away; this will, he believes, provide a metaphysical comfort for them, explaining that the chorus acts as ‘a living wall against the assaults of reality because it […] represents existence more truthfully […] and completely than the man of culture does, who ordinarily considers himself as the only reality’ (Nietzsche, 1993, pp.201-2). What Nietzsche means by this is that the chorus, and Greek tragedy as a whole, helps to bring forth matters such as death and one’s own mortality through the form of art, which in turn makes it a great deal more bearable. The metaphysical comfort springs from being able to consider these issues in an AP way, for the beauty of the dialogue and the poetry utilized within the play is logical and structured – AP indeed. It is this combination of AP and DI that Nietzsche admires, claiming it helps to ‘sugar the pill’ of unpleasant thoughts by displaying them in an aesthetically pleasing way.
It is when this careful combination of the AP and DI ceases that Nietzsche believes the artform to be in decline, a fault he places on Euripides. Euripides founded New Attic Comedy and was, Nietzsche claims, the first to bring the ‘spectator’ onto the stage. His plays, in contrast to that of Achilles or Sophocles, focused not on great tragic heroes but the ‘common man,’ dramatic reinterpretations of everyday occurrences similar to modern-day soaps. This led to the audience feeling they ‘knew’ the characters; that they could pass moral judgement on them in ways they could not on tragic individuals, and in turn threaten the return of individuation, as each spectator forms moral opinions on the events before them. Euripides’s plays were also carefully written and structured, much unlike the chaotic mess of the preceding tragedies: seemingly very un-DI characteristics. Nietzsche believes that the plays are left only with the AP, an artform that in no way helps us seek the aforementioned metaphysical comfort.
Nietzsche’s concepts of the AP and DI, along with the entirety of BT, were not well received initially: the author himself called the book ‘badly written, ponderous, embarrassing, image-mad and image-confused’ (Sweet, 1999, p.49). However, the ideas have left a notable legacy and are still discussed in matters of ethics, politics and art. Indeed, Michael Motta (1991) likens the opposing forces to mental illness, stating how stages of mania in bipolar artists can be likened to that of the DI (‘The urge to create is great, but the ability to step back, to control the process is reduced’), with depressive phases linked to the AP (‘Critique and reflection take precedence over impulsivity, inhibition holds sway over exhibitionism’). Nietzsche was also keen to pose these concepts politically, making great distinctions between the primitive DI state (a tribal, barbarian society), the apollonian state (disciplined and lawful societies), and the AP-DI state which he of course favoured (pp.29-36). However, Nietzsche considered the Greek culture to be a state of one of the two extremes – primarily AP – with only DI resemblance in certain ‘barbaric’ times across history. We can of course apply this to modern-day politics, though one could question its usage. Were we to accept Nietzsche’s ideas in relation to the structure of our own state, it would surely be that of the AP; and it would seem necessary to reconsider our government’s strict laws. Oswald Spengler (1991) echoed many of Nietzsche’s sentiments: in his book The Decline of the West, he shamed the ‘mass culture’ of modern western civilisation, decreeing it to be full of ‘clones’ created and moulded by the media, completely devoid of the DI aspect that both he and Nietzsche thought necessary for a thriving, illuminated culture (see ch.3-6). Spengler (1994), too, longed for a combination of DI and AP and the deconstruction of ‘bourgeois culture’ (pp.142-59). However, Nietzsche’s ideas seem to demand a certain degree of chaos and unpredictability in his AP-DI culture, one which the majority of modern-day democracies would be unwilling to accept, even when combined with the AP aspects we currently employ. Whether or not Nietzsche’s reasoning can be applied to the Greek culture to which he originally referred is difficult to ascertain; but it seems unlikely that such a literal usage of his ideas is useful in more modern discussions.
We can see the legacy of the AP and DI concepts even more clearly in psychoanalysis: Freud’s original theory is clearly grounded in the same ideology that Nietzsche employed, with the DI id contrasting to the AP superego. He praised Nietzsche as a forerunner of his ideas and took them a step further, universalising them to a topic relevant to every individual, whose minds could then continue battling the two extremes (Egels, 2000, p.96). Freud did, however, seem reluctant to concur with Nietzsche that the DI should be as equally utilised as the AP. It was not until Otto Gross that a psychoanalyst would fully appeal to the idea of a DI culture: Gross thought, in contrast to the idea that such a lifestyle would be destructive to society, that the progression and stability of a society depended on it. He spoke out for a ‘hedonist’ way of life, embracing unrepressed sexual activity and almost animalistic tendencies. It is unsurprising that Gross came to be considered an anarchist, and his ideas overwhelmingly dismissed by the world of psychological research (ibid. pp-301-8). It seems clear that in a modern society such ideas are not only impractical but potentially dangerous, undermining liberalism and current laws. While psychoanalyses have been hugely important in shaping psychological research and development, the influence of Nietzsche’s conceptions of DI and AP appears to be minimal outside of its initial inspiration of Freud’s id, ego and superego. Here, much like their influence on modern politics, the themes presented in BT, and in particular the supposed necessity of a DI aspect to culture, seem wholly incompatible with a modern liberal society.
Somewhat like Freud, Camille Paglia (1990) took Nietzsche’s ideas of the AP and DI to be biological, claiming that ‘the quarrel between Apollo and Dionysus is the quarrel between the higher cortex and the older limbic and reptilian brains’ (p.133). A dissident feminist, her thesis claimed that the DI was associated with women (and their hedonistic, chaotic lifestyle) while the AP is the male portion of society rebelling against their DI nature. It is due to this supposedly superior virtue that males have been dominant in many areas of culture, politics, science and history, with Palgia even going so far as to claim that ‘the male orientation of classical Athens was inseparable from its genius. Athens became great not despite but because of its misogyny’ (p.140). While it is important to note that such anti-feminist ideals were not that of Nietzsche’s, his influence on their creation is clear. Such notions in modern society would be, by the vast majority at least, considered not only insulting to women but going against the kind of liberal progression which has taken place over the past few centuries. So the concept of the DI and AP can be seen to not only endanger the political world but the social one, and, once more, to be wholly incompatible with today’s western culture.
There are, however, certain contexts in which Nietzsche’s concepts are relevant and potentially illuminating. Written as they were about Greek culture, we can perhaps hope to gain some insight into the structure of ancient Greek society, and in particular their relationship towards art and theatre. More recently, the concepts have been used anthropologically by Ruth Benedict (2009) to illustrate the differing values of varying cultures, such as the self-restraint of the Zuni people, which would be associated with the AP, and the somewhat vulgar and ostentatiousness of the Kwakiutl, associated with the DI (pp.130-51). We have seen, through its implications and impracticalities, that it would be difficult and dangerous to implicate the concept of a culture that employs both DI and AP lifestyle in modern western cultures; but perhaps, as with the above, they are useful in our exploration of ancient and/or anthropological societies. The opposing forces are certainly a crucial matter in the study of Greek art and continue to be an interesting line of thought when studying cultures outside current modern liberalism.
Benedict, Ruth, ‘Configurations of Culture in North America’, American Anthropologist 34 (American Anthropological Association, 2009)
Nietzsche, Friedrich, The Birth of Tragedy (Penguin ed., 1993)
Paglia, Camille, Sexual Personae (USA: Yale University Press, 1990)
Sedgewick, Peter, Nietzsche: The Key Concepts (Routledge, 2009)
Spengler, Oswald, The Decline of the West (OUP Australia and New Zealand; abridge edition, 1991)
Sweet, Dennis, ‘The Birth of The Birth of Tragedy’, Journal of the History of Ideas, Vol. 60, No. 2 (Penn Press, 1999)
City College of New York
A long-debated problem in the philosophy of artificial intelligence concerns whether intelligent machines are able to understand their own tasks. Some intelligent machine-engineers and researchers might say that whether a machine understands the tasks it is performing has no implications as to its pragmatic function, which concerns how efficiently the machine can execute these tasks. But for certain tasks an intelligent machine must surely understand the tasks it is performing. An ‘understanding machine’ may be the only kind of machine that stands a chance of passing the Turing Test: an understanding machine would be able to consistently compose meaningful sentences and thereby communicate efficiently—the way humans communicate.
An understanding machine might even be able to produce brilliant works that rival renowned writers like Shakespeare. I say this because language is determined by both semantics and syntax, if the machine does not possess any semantic properties it must rely on syntax and specific software that correlates words. Such a machine would only be able to compose a limited number of intelligent sentences, however, and this would be the equivalent of a human manipulating the words of a language he or she does not understand. Considering this it seems a little far-fetched to expect a syntactic machine to consistently compose meaningful sentences. An intelligent human who does not have knowledge of a new language would be incapable of so composing consistently intelligent sentences, and it seems that the expectation of syntactic machines to perform such a task is not grounded in a reasonable assumption.
The assumption is that syntactic machines are able to communicate intelligently without understanding a language but the standard we use for comparing intelligence is that of humans, for there is no further recourse. If an intelligent human cannot compose meaningful sentences without understanding the language at hand then by what logic are we to assume that an intelligent machine could? The assumption completely neglects the important function of understanding and semantics in intelligent communication, which is obviously evident in humans.
Even though the mechanism behind how semantics and understanding emerges in human cognition is still poorly understood, one cannot say that the mechanism that allows humans to understand language is irrelevant in intelligent communication. If the Turing Test is a measure of artificial intelligence, machines that understand language would surely have the best chance of passing the Turing Test. The problem of engineering such a machine becomes apparent when one delves into the nature of meaning and the emergence of meaning from electrical impulses—though there are certain human attributes that an intelligent machine needs to possess in order to understand language.
2. The Emergence of Meaning
In the ‘Chinese Room’ thought experiment John Searle proposed that artificially intelligent machines, even if conscious, would be unable to understand language on account of the way they function. Computers manipulate symbols and words based on syntactic instructions pre-written in their programs. The computer is following instructions, and though it does this very well, there is no human learning or understanding involved in the computer’s processing these instructions or its generating outputs. It seems that it is hard to comprehend how an intelligent being begins to understand language; and this is of course expected since the understanding of language relies on the emergence of meaning. One has to understand the meaning of words in order to understand language. The hurdle of engineering a computer that understands language is concerned with the emergence of meaning.
There is no plausible mechanism or an objectively satisfying explanation that describes how meaning emerges from electrical impulses. This applies whether we are talking about the electrical impulses in a functional computer or the electrical impulses generated by human neurophysiology. As humans we know things mean something to us; we have meaningful experiences but such are not reducible to our neurophysiology.
John Searle demonstrated that the emergence of meaning is not reducible to the instructions coded in computer software when he created the ‘Chinese Room’ thought experiment. Searle showed that a computer which is efficient at generating language could never apprehend the meaning of the words and sentences it is generating. Searle asks the reader: At what point in this mechanism, which is modelled after typical computer software, does meaning emerge? The conclusion is obvious: if the machine only follows instructions and never grasps the meaning of words, it will never learn what they denote and as a consequence it will never be able to understand language.
Similarly we can ask where meaning emerges in our neurophysiology. Can we say that a single neuron can have a meaningful experience when it is stimulated by electricity? What about a small network of neurons? We can then further extend the network to include millions or billions of neurons and we can still ask the question of emergent meaning: when are these neurons capable of having a meaningful experience? We can infinitely add more neurons and make the neural network infinitely more complex; but it appears we will never know how meaning emerges from neurons, since their number or complexity of network provides us with nothing by way of a plausible mechanism for the emergence of meaning. We can always code more instructions into the computer software and make it infinitely more complex, as Searle proposed, but it seems semantics will never emerge from this.
Although how meaning emerges from a physical or a biophysical mechanism is a puzzling and confusing question, we can think about how the meaning of words begins to emerge during development. We can take a simple example from our everyday experience and observe how children learn new languages. Children learn to speak before they learn to read; they do not have the convenience of consulting dictionaries. In truth we learn very few words by consulting dictionaries, and if we think about the emergence of language in general, it must have originated before words were institutionally defined: we learn language by associating words with experiences, phenomena, and concepts.
The way people understand words is also highly abstract. Understanding does not emerge solely within associations between objects and words; it emerges as a web of associations between a concrete object, its abstract representation (which does not have a substance or form), and the term that describes the object. This abstract representation is what we call conceptual understanding. Our conceptual understanding of a tree, for instance, is not represented by any specific physical example of a tree; it is represented by our abstract understanding of it, understanding as a collection of properties or information that pertain to specific trees.
This web of associations is crucial. If we associate a word and a concept but fail to associate both word and concept with an actual object, we are stuck with abstract definitions which would seem like a collection of information that does not apply to anything specific or actual. In such a case we could learn about the word ‘tree’ and what the properties of a tree might be, but we would not recognize a tree if we were to come across one. On the other hand if we associate the word and the actual object, but fail to associate the word and the actual object with the concept, we would then be unable to collectively identify several different types of trees, or a picture of a tree, even though we are able to identify a specific. The understanding of what words mean requires a web of associations between concepts, objects, and words; an intelligent machine intelligent enough to understand language must thus possess this ability to create webs of associations.
3. The Problem of Precise Definitions
The meaning of a word is associated with a the phenomenon it denotes. We can, however, semantically associate a given word with other words that describe or define the word in question. We often attempt to provide a precise definition of a word in terms of others: for instance we can identify the word ‘tree,’ the object, and the concept with which it is associated, but we cannot describe what the word actually means, precisely, unless we use other words to provide an accurate description. We might say the tree is a living organism, it has branches, or it has leaves, and so on.
Language acquisition through precise definitions is impossible, however—since if we fail to associate words with anything else, other than words, we must be stuck in a cycle of ignorance (which we are not). If we start looking for the precise definition of any given word, we have to understand it in terms of other words, but since we do not understand what these others mean we have to look for their respective definition as well. We look endlessly for definitions of words we do not understand, and if we fail to associate these with anything beyond lingual terms, we fail to understand what any of those words might mean.
We may think of a given word W, and this word is precisely defined by another set of words, W1. But because each of these words in W1 must also be defined precisely, there has to be yet another set of words, W2, which defines those in W1. And again, since the set of words in W2 require precise definition, there must be a further set of words, W3, that defines those in W2—ad infinitum.
The problem here is that once we start to define words in terms of other words we end up going into an infinite regress of definitions in order to find the precise definition of any given example. We define endlessly without ever hitting that precise definition. This is why words are naturally vague and ambiguous: they cannot be defined precisely.
So it is clear that one cannot initially learn the meaning of words in terms of others, since this kind of learning is impossible as demonstrated above. Someone who attempts to learn the meaning of words solely by rifling through dictionaries falls into an infinite regress of definitions which appear ultimately meaningless to the reader. Language has to be learned through associations of words with concepts and actual objects. The intelligent machine that understands language must possess some kind of sensor that allows it to make associations between words and objects. Computers can easily make these associations for they are simple relations between concrete symbols and objects. It is difficult, however, to know whether the intelligent machine is able to form a concept from this association. The major problem with conceptual formation is that the process requires semantics. It is a mental process that requires a certain understanding in order to make successful associations between concept, word, and object. This understanding relates to how and why a collection of properties, or information, pertains to a specific object.
4. The Learning Machine
Learning is crucial for language semantics. From what we know about language acquisition, one cannot develop the semantics and the understanding of language without first learning that language. Previously I mentioned why learning is crucial for language acquisition and why language needs to be learned by the association of actual objects with concepts and words. The process of learning a language is also the process which allows us to begin understanding language. Language acquisition is therefore developmental by nature: it is something that must develop through an interaction of organism and environment.
Learning the meaning of words through association with others is problematic. Such a method of learning does not allow for understanding on account of the reasons stated above. Similarly we can conceive of a Turing machine bearing database of all words, books, and encyclopedias ever written: this machine has an endless supply of literature; it has enough computing power to associate words with other words and words with sentences; and from these correlations it is able to compose intelligible discourse. But such a computer would be unable still to understand the meaning of the words it uses. It can only correlate words with other words, and words with sentences, but under these conditions semantics and understanding cannot emerge. The machine knows how words correlate but it cannot know how they are correlated to the phenomenon they are intended to denote.
What if we installed a sensory device and we programmed the machine to associate words with objects? In such a case the computer might make intelligible associations between such things, but as stated above the machine must also have a conceptual understanding of the words it uses. Nor can organisms that understand language rely on a static, pre-determined code to learn language because one cannot program the inductive knowledge necessary for language acquisition using a static pre-written code. The machine might have an infinite number of terms and sensory devices installed but if it cannot encounter the actual phenomena the words are meant to denote it cannot have perception and understanding of what those phenomena might mean: it cannot understand how the words are related to existing objects. It lacks the capacity to acquire inductive knowledge.
If we posit that a machine can understand without acquiring inductive knowledge it would be similar to stating that the machine can somehow know or understand facts about the world without encountering or observing them. It seems a little ridiculous to claim that a machine can acquire such ‘magical’ or ‘psychic’ knowledge, especially when discussing the kind of knowledge necessary to language acquisition.
A machine that runs on a static pre-written code is incapable of acquiring inductive knowledge because upon observation or encounter of a particular fact the machine must undergo changes in processing or operation. The state of the machine has to transition from ‘not knowing that particular fact’ to ‘knowing that particular fact.’ Learning a language requires some degree of creative self-processing because learning something new entails the altering of one’s processing. Such changes are regularly observed in organisms that have the capacity to learn and acquire inductive knowledge—not machines.
I will admit that the term ‘creative self-processing’ is somewhat vague. A creative self-processing machine would, for instance, alter its own pre-written processing—and we could always wonder at what point a machine, or its environment, might be altering its own processing. But can we say that a machine that evolves and acquires linguistic skills through association of words and concepts with its environment is learning in a human way? The machine does not have a pre-written program for language, but it picks up language as it interacts with its environment and the speakers it encounters. We could say that such a machine is ‘learning,’ but if it is doing so by association can we assume that the machine understands? We might assume that it understands but that this interpretation of ‘understanding’ is inaccurate for a number of reasons (and I will go into detail about this later).
Other important aspects in human learning are related to the plasticity of the brain. During development and learning the brain’s connectivity undergoes various changes, new connections, and the formation of new synapses. We know that as new neural synapses are being formed this phenomenon will alter the activity of individual neurons. Within the human brain we are not only concerned with decentralized processing, but also the fact that the processing units so applied tend to change as the human brain learns. There are myriad factors involved in human learning, such as epigenetic factors, which alter gene expression in the neural cells—once again altering the nature of the processing.
I would like to emphasize that human learning is a dynamic system where constant changes occur in the processing and these changes have an effect on how humans come to learn and understand language. These kinds of changes are the kind of changes that we would expect a creative self-processing machine to undergo.
5. Is the Learning Machine an Understanding Machine?
Previously I stated that association is necessary for understanding languages, but in this I did not mean it is sufficient: learning is required as well. Now we can ask if learning by association is sufficient for understanding. Once again the answer is probably ‘no.’ We know that both learning and association are important to language acquisition, but we are still a long way from making sense of the understanding machine. The emergence of meaning still remains a mystery. Learning by association does not guarantee that a conceptual understanding (the kind of understanding required for language semantics) will emerge. As a consequence we cannot at this stage know when a machine begins to form concepts. Conceptual formation is the necessary element of artificial intelligence that remains vague and poorly defined in the physical sciences. One cannot define a concept or the emergence of a concept physically: as I mentioned earlier the associations between concepts and words requires understanding. When one understands the concept of the word, then the meanings of that word begin to emerge.
Other difficulties that pose obstacles for an understanding machine are related to the means of perceiving phenomena. We know, for instance, that there are computers that play chess and possess learning algorithms but is it reasonable to claim these computers understand what chess is? Let us assume for argument’s sake that these machines have a conceptual understanding of it: but we must now think about all the other things the machine must understand in order to grasp chess the way we do. I do not think we can claim that chess-learning computers have an adequate understanding of chess on account of their lack of a wide enough understanding of the cultural context in which chess is played and the context in which chess is defined as a leisure game or a sport. The chess-playing computer cannot therefore understand chess, or rather it cannot understand chess the way we understand it. For us, however, understanding appears to emerge from perceiving an integrated network of phenomena and experiences. A machine that lacks such perception will be unable to understand things the way we do; and it would be ridiculous for us to expect the machine to understand phenomena in such a way if we do not share a common means of perception with it.
If we want to build a machine that understands language, we have to ask ourselves what else the machine must understand besides language. Like chess, language is probably not an isolated phenomenon; it is integrated with many other aspects of our phenomenal experience which we must ourselves understand before we can start developing our understanding of language. Humans must have a certain mechanism for perceiving language; and in order for an intelligent machine to understand it the way we do, it must share with us a common perception it currently lacks.
To summarize I have identified two properties which are crucial in order for the machine to understand language. One is the process of creating a web of associations between words, objects, and concepts; the second is the machine’s capacity to learn by association or, in other words, to acquire inductive knowledge. The difficulty in building a machine that understands concepts presents a major hurdle in building engineering something that grasps the meaning of language, and phenomena in general. Building a computer that understands concepts and knows the meaning of words is currently beyond the scope of scientific engineering, since there is no plausible physical mechanism that explains the emergence of meaning. Finally, the understanding of language requires a more integrated phenomenal experience and a human perception of language—none of which intelligent machines currently possess.