Originally published in Kristan Horton’s Oracle, 2001.

Writing Down What Is Said

It  could   be  said   that   we  are   witnesses  to  a technological revolution.  To some  this revolution is a liberating force, to others it is a threat to humanity, but  to most  of us it simply  represents a series  of bewildering changes to  almost  every aspect of our  lives.   The  reaction from  artists is equally  varied.   The artist  Kristan  Horton,  for one, has  thoroughly embraced the  revolution, and  has wasted no time in incorporating technological innovations into  his art  production.  Nonetheless, Horton has not been  intoxicated by it; which  is to say, he has not succumbed to the allure of applying technology in art for its own sake.  Instead, he has managed to remain aloof in this  regard.  As such he  plays  the  traditional role  of artist  as  outsider by  exploiting  the  technology  to  help  us  see  it anew.  His Oracle project typifies  this approach to art.   Below I shall discuss some  of the  issues  that this project raises  concerning the nature of the relationship between speech and writing.


A physical description of Oracle is straightforward: the  Oracle  machine comprises a row  of six-tape players each plugged into a relay switch, which in turn is connected to a computer. Once the first tape has  finished the  machine is programmed to  cue the next tape-player to start while simultaneously switching to  allow  the  computer to  receive   the audio   signals   from  this  next  tape-player.    This process continues until all six tapes have been played,  at  which  point  the  machine loops  back to  the  first  tape  to  start the  process over  again. Installed on  the  computer is  dictation software, that is, a program designed to interpret the spoken word – converting the audio signals  into text.  This text is printed off, its pages to be eventually bound. In addition a monitor continually displays the interpreted text as the tape plays.

All six  tape-players are  used  to  play  a nine-hour recording of a revised version of Homer’s Odyssey read by Alex Jennings1.  The choice of this story  is centrally relevant to the Oracle project. Homer  lived in preliterate Greece, and so he was a  storyteller in  the  oral  tradition.   Accordingly, his  stories i.e., The  Iliad  and  The  Odyssey,  were passed on  to  the  later  classical  Greeks  by  word of  mouth, only  then  to  be  preserved in  written form.  The advent of the alphabetic writing  system in  Greece   in  many   ways   was  a  revolution as profound in its consequences to everyday life as today’s technological revolution.  And like today’s technology, writing  was  not  favourably received by everybody.  One of writing’s most renowned opponents,  ironically,   was  the  philosopher Socrates2. His argument against writing principally hinges on the  claim that  it lacks  vitality.   That  is, he believed that  writing  petrifies the thoughts expressed in the  speech it records. While  writing can  record or  preserve the  fleeting ideas  which we both  utter and  think,  it lacks  the  immediacy of speech which  is the  quintessential medium of thought3.  Of course this belief betrays the primacy of speech which  Socrates presupposed, in that  he saw writing  merely  as an instrument for recording speech, rather than  a medium in which  vital communication can itself occur.   This belief in the primacy of speech surely derives from the fact that we master language through speech – writing  is an addendum.

Writing   is  essentially  mechanical, and  it  is  this fact which explains how it has become such  an integral  part   of  our   technologies,  in  particular in how  we interact with  computers.  Speech,  by comparison, is unwieldy.    To an  extent we have compensated for this lack of immediacy in our writing  based  technologies by exploiting the exponential speed  of electronic media.    And this compensatory processing speed  has  led us to try to give to computers the  power  of speech which they   so  conspicuously  lack,  or  more   precisely in this  case  the  power  to recognize speech.  It is these attempts which  are  the  background to the Oracle   project.   From  a  certain  viewpoint one can  read  Horton,  through Oracle,  to be mocking such  attempts: The text  produced from  the  tapes is almost  wholly  unintelligible.   For example,  we get  “implicit to  dissuade unsealed  slide  to  go down  to  laugh   in  disguise…” and  so  on.  The program  used   by  Oracle   i.e.,  IBM’s  ViaVoice, can be ‘tuned’ to produce a more accurate interpretation of the taped reading. But instead, by leaving the interpretation loose, Horton could be said to be highlighting the gap between the mediums of speech and writing. That is to say, he points to the essential difference between them which we attempt to overcome by sheer computer processing power. He remarks that such speech recognition programs represent an impulse to turn books-on-tape back into books which, he notes, is ridiculous from the technology industries’ perspective. Of course, dictation software in general, and ViaVoice in particular, is not intended for this particular purpose. But, that might be partly why he chooses to perform this task with it – to show the futility of this whole project. After all, it can be argued that the computer does not so much recognize speech, but rather converts it into a mechanical medium, namely writing which, as a machine, it can process. In other words, dictation software is really designed to mechanize speech.

As mentioned, the choice of Homer’s Odyssey is central to the Oracle project. For centuries, before it was eventually preserved in its written form, this story existed in virtue of being memorized by generations of storytellers i.e., bards. Moreover, the Odyssey is often referred to as an epic poem because it was recounted in rhyme in order to aid its tellers in their memorizing of it. Compared to ourselves, members of oral societies possess astonishing capacities to remember the spoken word, and rhyme is one of their primary aides-mémoire. Perhaps only with children’s rhymes recited in schoolyards does this oral tradition survive in our society. Notably, the word recognition program ViaVoice used by Oracle also exploits rhyme, or more precisely sound patterns in speech, to help it interpret the spoken words. Thus, its interpretation is never entirely arbitrary, that is to say, it does not simply interpret each word individually – sound patterns at least give the program a foothold in interpreting each word in some sort of context. So, for instance, the words “there’s a distant sound” might be interpreted by Oracle as “the airs a decent hound”. In this sense Oracle’s approach is eerily human, producing variations of the Odyssey in similar manner to how the ancient bards also produced variations of the story through the generations, echoing this oral tradition. Horton observes that this approach produces a reversal of the intended process of writing generally, which is to harden the ideas expressed in speech. What we in fact get is a loosening of what is spoken, namely the Odyssey according to its preserved written form as read by Alex Jennings. The machine becomes poet – an interpretative creator of texts.

Yet, crucially Oracle’s creativity lacks any semantic dimension. That is, while the computer converts human speech into writing, its dictation software is not at all concerned with the content of this writing. One might say that Oracle is blind to linguistic content in general. That is why its interpretation of the tape is often unintelligible – it interprets each spoken word without any regard for its meaning. For us writing down what is said involves being able both to mechanically write out the signs which represent the spoken words and to comprehend those words – Oracle can do the former, but not the latter. Given that poetry is essentially semantical in nature, that is, its vitality derives from its meaning, we see that Oracle’s output merely resembles poetry.


So far I have suggested that Oracle’s activities are a charade. That is, at first blush Oracle creatively interprets human speech, whereas in fact, its activities merely seem creative – underneath there is only an empty mechanical process. Yet, this reading of Oracle seems too harsh. Indeed, were one to watch Oracle’s monitor actively generating text one would likely think that there is genuine thought occurring – one would feels that something is going on inside that computer, however primitive and mechanical its processes might be in comparison to humans. Further, Oracle   possesses  an   autonomy,  albeit   limited, which is surely  the defining feature of a minded being.  This autonomy is evidenced by the fact that its  textual  output is never  the  same  – each  time it returns to the  same  passage it offers  up a new interpretation.

The suggestion that  machines can be minded, or even  are  so, is much  debated.  The English   mathematician  Alan  Turing4   contended that  if a machine can behave in a manner indistinguishable from  intelligent behaviour, then in effect that machine has a mind in this regard. In reply to this contention the American  philosopher John  Searle5 presents an  argument to show  that such outward resemblance to minded behaviour is not evidence for genuine mindedness. Briefly, Searle  imagines a man  locked  in a room.   In the room the man has a book which lists a set of rules indicating how  to reply  to any  string of Chinese characters using  Chinese  characters written out on individual cards.   The man, or operator, speaks no Chinese  and so relies  entirely on the rule-book to  reply   to  Chinese   sentences passed through an opening into  the  room.   Thus,  the  man  offers intelligible replies  to  such   sentences  using   the cards,  which  he  posts  back  through the  opening a  while  later.    One  might  conclude that  to  the outsider the operator in the room understands Chinese. This  claim  is parallel to  Turing’s concerning intelligent behaviour. But  Searle observes that  this  operator does  not  understand Chinese,   and   therefore  the   inference  that   he does from the evidence of his replies is invalid. Likewise, even  if a machine like Oracle  were  not only able to accurately interpret speech, but could also manipulate words  in a syntactically correct manner, call it super-Oracle, still this would not be evidence that  it is in any sense conscious.  Super-Oracle would merely  seem minded.

However, Searle’s Chinese room is not comparable to  the  kind  of  machine with  which we are  concerned i.e., a super-Oracle.    The programs of  a  super-Oracle would  perform the same  function as  the  Chinese  room’s  rule-book, but notably without an operator who can or cannot speak  Chinese.   Accordingly,  for such  a computer its apparent ability  to understand Chinese  cannot be refuted by pointing out that its operator cannot in fact understand a word of Chinese, since no operator is involved.    Further, more  generally it can be argued that  the  fact that  Searle’s  operator cannot speak  Chinese  is wholly irrelevant, it is the ‘room’ which   ostensibly understands  Chinese. The  fact   that   the   operator  cannot  understand Chinese  does  not  refute the  claim  that  the  room can.  Still, we can sympathize with Searle’s  overall claim  that  understanding does  not  reduce to  an ability to manipulate signs i.e., a mastery of syntax – even a super-Oracle cannot surely  be said to understand the language it operates on.

There  seems  to be one crucial  difference between Oracle  and the Chinese  room: there is no rigid connection  between  Oracle’s   input   and   output – as noted,  each  time Oracle ‘hears’  the  same passage  it  produces  a  new   ‘interpretation’.    In the case of the Chinese  room, even though it is operated by  an  autonomous  human, its  output is determined by  the  rule-book, and  accordingly it does  not  produce a new response each  time  to the  same  input.   Perhaps, however, this rule-book could  be  written in such  a way  that  the  human operator does  have  a  choice   of  words,   and  so the  Chinese   room’s   output would   vary  for  the same   input.   Nonetheless, still  its  operator has no relevant criterion by which  to make  a choice between words – he operates blindly.   In Oracle’s case, its interpretation i.e., output, is governed by how well it has been trained – the more thoroughly trained Oracle  is the  higher the  probability that it always  produces the  same  interpretation each time  a  passage is  repeated to  it.    No  training is  involved   in  the  production  of  the  rule-book for  the  Chinese  room;  its  rules  are  given,  so  to speak.   And in this respect Oracle is more human. Oracle’s  training mimics  how  we learn  to speak, and  later  write;  namely  by a constant process of trial and error.  In this sense,  Oracle  is not purely a machine.   It is important to  understand what is  meant by ‘machine’ in  this  respect.   Here, by machine I mean  any  physical object  which  is designed to perform a specific  task.   In order to perform its task  all of a machine’s operations are predetermined.  In this respect the  Chinese  room is essentially a machine because all of its input  to output relations are predetermined, even if its rule-book is written so that its operator has a choice of words  – still this  choice  is strictly predetermined. Oracle,  by contrast, is less of a machine because its input  to output relations are not entirely predetermined.

That  aspect of Oracle  which  is not  machine-like i.e., mechanical, is instead organic.    Thus,  Oracle is in part  an  organism.  Of course, ordinarily an organism is defined as a living entity.  However, as a physical object, I suggest that an organism differs from a machine in terms of how it is designed. As noted,  a machine’s operations are predetermined; by  contrast,  an  organism’s  operations  are determined in relation to its environment.  Thus, for  example,  we  say  that  a tree  has  an  organic form,  because its  form  has  been  determined by its environment – the prevalent direction of the wind,  whether it  stands on  flat  of inclined  land etc.   Likewise, an airplane has a mechanical or inorganic  form   because  it  is  designed  with   a specific function in mind.  Accordingly, we see now why speech is essentially organic while writing  is not.   Speech  arose  from  our  interaction with  our environment, while writing  was designed by us to perform a  particular function, namely  to  record what  is said.   The  fact  that  speech is constantly evolving   is  evidence of  its  organic nature  – its design  alters with changes in the environment, namely  those  new aspects of the environment we find need to communicate to others.

Thus,  in order for  a computer, such  as  the  one used  by  Oracle,  to  master speech it must  learn to  recognize speech in  an  organic way,  namely by allowing  its operations to be partly determined by its interaction with the  environment; which  in this  case  is specifically the  degree of correctness the  computer achieves in its interpretation of the spoken word.   By not training the computer to accurately or  correctly interpret  speech, Horton effectively  emphasizes the  organic nature of Oracle’s  design  in this  regard.  It is this  organic design,  I suggest, which  encourages us to ascribe creativity to Oracle’s operations. In this sense, talk of whether machines such as Oracle are minded or not does  not help  us to understand the  crucial distinction between machines and organisms. Moreover, not all organisms are minded, and indeed among  the plethora of organisms presently existing, we surely ascribe to them varying degrees of  mindedness.   It  is  Oracle  as  an  organism, rather than  as a putative minded being,  which is relevant in explaining its creativity.    Can we therefore admit  a  degree of  poetic  creativity to Oracle?   I stated that  poetry per se is essentially semantical,  and  Oracle’s   programming  is  blind to the meaning of the speech it interprets. Nevertheless,  while   we  might   not   be   able   to describe Oracle’s  output as  poetry, inasmuch as rhyme is an aspect of poetry Oracle  is indeed creative.  And this creativity derives from Oracle’s organic nature.

© Hugh Alcock

  1. The Odyssey read by Alex Jennings (audio tapes), London: Penguin Books.
  2. Socrates’ views on writing are clearly laid out in Plato’s dialogue Phaedrus (see especially 274-279).
  3. Roughly, according to Socrates, speech, as the medium of thought, is the vehicle for the acquisition of knowledge i.e., the truth. Moreover, dialectical debate i.e., a reasoned inquiry through the discussion of specific concepts, can by itself provide knowledge. A famous example of such an inquiry features a discussion between Socrates and a boy slave of a young aristocrat named Meno. Without teaching the slave the concepts of geometry – which the slave has never learned – Socrates seems to succeed in demonstrating that still the slave can recognize for himself certain truths of geometry in virtue of his discussion with Socrates. (see Plato’s dialogue Meno 81b-8
  4. Alan Matheson Turing is most famous for his achievement of working out the Enigma code machine used extensively by the Germans to send secret messages during world war II. His views on consciousness are found in his article ‘Computing machinery and Intelligence’, Mind, 1950.
  5. See Searle’s article ‘Minds, Brains, and Programs’, The Nature of Mind, David Rosenthal ed., pp. 509-519, Oxford: Oxford Univ. Press, 1991. The Chinese room

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s