Weizenbaum's Secretary

Reconstruction of CACM 1966 conversation (2024). 

Alternative Secretary Conversation 22 March 1965

Example of an alternative conversation (22 March 1965) MIT Archives (Computer Conversations 5 of 7). 

from Project Mac (1966) Project Mac Progress IV 1966-67,  p. 126.

From Weil, A. (1965) “Conversations with a Mechanical Psychiatrist.” The Harvard Review, Vol. 3, No. 2, pp. 68–73, 1965.

From Weizenbaum, J. (1967)

20 Mar 2024

Since 2021, part of the work of the Eliza Archaeology project has been trying to identify who might be the author of the conversation that Weizenbaum referred to in his 1966 paper and again in Computer Power and Human Reason (1976). He writes:

I was startled to see how quickly and how very deeply people conversing with DOCTOR became emotionally involved with the computer and how unequivocally they anthropomorphized it. Once my secretary, who had watched me work on the program for many months and therefore surely knew it to be merely a computer pro­gram, started conversing with it. After only a few interchanges with it, she asked me to leave the room... And I knew from long ex­perience that the strong emotional ties many programmers have to their computers are often formed after only short exposures to their machines (Weizenbaum 1976: 6-7).

As Weizenbaum goes on to explain,

The following conversation be­tween a young lady and Eliza playing doctor illustrates both the Rogerian technique of encouraging a patient to keep talking and the operation of the computer program ELIZA. The first to "speak" is the young lady. The computer's responses are printed entirely in capitals.

Men are all alike.

IN WHAT WAY

They're always bugging us about something or other.

CAN YOU THINK OF A SPECIFIC EXAMPLE

Often it seems that the "secretary" referred to in Weizenbaum (1976: 6) has been inadvertently conflated with the "young lady" in the conversation (Weizenbaum 1976: 3), particularly by later writers. But the question is are they actually different people? Additionally, who are these unnamed people ("secretary" and "young lady") whose conversations are recorded with ELIZA in Weizenbaum's work? 

In our research team we have discussed whether this person(s) might be fictional or a composite of different users of the ELIZA system. Indeed, we have found different versions of the conversation with the "young lady" in the archive and hand-edited records that indicate that the published version was tidied up before publication (see image on left, Alternative Conversation 22 March 1965, but see also Weizenbaum 1965). For example:


Weizenbaum (March 22, 1965, MIT archive)

I AM SORRY TO HEAR YOU ARE DEPRESSED

He's right. I am unhappy.

DO YOU THINK COMING HERE WILL HELP YOU NOT TO BE UNHAPPY

Perhaps I could learn to get along with father.


Weizenbaum (1965)/Project Mac (1966)

I AM SORRY TO HEAR YOU ARE DEPRESSED

he's right. i am unhappy

DO YOU THINK COMING HERE WILL HELP YOU NOT TO BE UNHAPPY

i need some help, that much is clear


Weizenbaum (1966: 36-37)

I AM SORRY TO HEAR YOU ARE DEPRESSED

It's true. I am unhappy.

DO YOU THINK COMING HERE WILL HELP YOU NOT TO BE UNHAPPY

I need some help, that much seems certain.


Weizenbaum (1967: 475)

I AM SORRY TO HEAR YOU ARE DEPRESSED

It's true. I am unhappy.

DO YOU THINK COMING HERE WILL HELP YOU NOT TO BE UNHAPPY

I need some help, that much seems certain.


Weizenbaum (1976: 3-4)

I AM SORRY TO HEAR YOU ARE DEPRESSED

It's true. I am unhappy.

DO YOU THINK COMING HERE WILL HELP YOU NOT TO BE UNHAPPY

I need some help, that much seems certain.

This version variation in published and unpublished versions of the ELIZA output is not commented upon in the literature, which is curious in itself, but it remains likely that Weizenbaum drew on multiple conversations to create them. There are other variation differences in the texts, e.g. "mother" is not mentioned in the archive 1965 version and there are different responses to the line "DOES IT PLEASE YOU TO BELIEVE I AM AFRAID OF YOU". By 1966 the text appears to stabilise and remains the same in subsequent versions. However, in all cases, the name of the person (or another user or users) remains unknown. Indeed, Weizenbaum variously refers to the person as "young lady" (Project Mac 1966: 126) or "distraught young lady" (Weil 1965), "user" (Weizenbaum 1966, 1967), "patient" and "young lady" (1976). 

It is notable that a similar conversation is given in McGuire, Lorch, and Quarton (1967: 180). Weizenbaum notes in an interview with McCorduck (1972) that he met Gardner C. Quarton, who was professor of psychiatry at the Harvard Medical School and particularly at Mass General Hospital (MGH) whilst he was developing ELIZA. Quarton "took an interest in all that and so I started working with Mass General" and it may have been through this link that some of the conversations that Weizenbaum uses, or assembles, in his publications are generated from this connection.


McGuire, Lorch, and Quarton (1967: 180)

Subject: I have been very depressed.

Response: Would you tell me more about these feelings.

Subject: They usually start whenever I go to work.

Response: Could you elaborate a little bit.

Subject: Well, it just happens that.

Nonetheless, it is clear that Weizenbaum's secretary actually used the ELIZA system. As he writes in Weizenbaum (1967: 477),

My secretary watched me work on this program over a long period of time. One day she asked to be permitted to talk with the system. Of course, she knew she was talking to a machine. Yet, after I watched her type in a few sentences she turned to me and said "Would you mind leaving the room, please?" 

Weizenbaum writes about the effect, later actually coined the "Eliza Effect", that this dyadic conversation could have between user and computer, 

What I had not realized is that extremely short exposures to a relatively simple computer program could induce powerful de­lusional thinking in quite normal people. This insight led me to attach new importance to questions of the relationship between the individual and the computer, and hence to resolve to think about them (Weizenbaum 1976: 6-7).

The Eliza Effect is the propensity for humans to ascribe understanding and intelligence to computer systems. Hofstadter (1995: 167) described it as “the susceptibility of people to read far more understanding than is warranted into strings of symbols - especially words - strung together by computers”, a compelling description written in 1995 but which accurately describes generative AI systems today like ChatGPT. Similarly, Turkle described that “the Eliza effect refers to our more general tendency to treat responsive computer programs as more intelligent than they really are. Very small amounts of interactivity cause us to project our own complexity onto the undeserving object” (Turkle 1997: 101). 

However, more recently one of our informants who worked with Weizenbaum in the 1980s claimed to have been introduced by Weizenbaum's secretary to the ELIZA system when visiting MIT. The informant even claims that this was the original "young lady" and that Weizenbaum made a joke about her use of the system. However, unfortunately, as yet, we have been unable to pin down an actual name. 

To try to resolve this issue I’ve tried to cross-check with other records in the MIT archives but I haven’t found anything substantive yet. These are, I think the most likely people, if it was a secretary. The following are the people who were support staff in Project Mac whilst ELIZA was being programmed/released. 

Blog Post by David M. Berry

Appendix I

Project MAC Administrative and Supporting Staff (1963/64)

Project MAC Administrative and Supporting Staff (1964/65)

Project MAC Administrative and Supporting Staff (1965/66)

Project MAC Administrative and Supporting Staff (1966/67)


* in post throughout the period of Project Mac 1963-1966

** in post throughout the period of Project Mac 1964-1966


Appendix II

The text of the exchange between the unnamed "young lady" and ELIZA is as follows, the computer's responses are printed entirely in capitals (Weizenbaum 1976: 4-3).

Men are all alike.

IN WHAT WAY

They're always bugging us about something or other.

CAN YOU THINK OF A SPECIFIC EXAMPLE

Well, my boyfriend made me come here.

YOUR BOYFRIEND MADE YOU COME HERE

He says I'm depressed much of the time.

I AM SORRY TO HEAR YOU ARE DEPRESSED

It's true. I am unhappy.

DO YOU THINK COMING HERE WILL HELP YOU NOT TO BE UNHAPPY

I need some help, that much seems certain.

WHAT WOULD IT MEAN TO YOU IF YOU GOT SOME HELP

Perhaps I could Jearn to get along with my mother.

TELL ME MORE ABOUT YOUR FAMILY

My mother takes care of me.

WHO ELSE IN YOUR FAMILY TAKES CARE OF YOU

My father.

YOUR FATHER

You are like my father in some ways.

WHAT RESEMBLANCE DO YOU SEE

You are not very aggressive but I think you don't want me to notice that.

WHAT MAKE YOU THINK I AM NOT VERY AGGRESSIVE

You don't argue with me.

WHY DO YOU THINK I DON'T ARGUE WITH YOU

You are afraid of me.

DOES IT PLEASE YOU TO BELIEVE I AM AFRAID OF YOU

My father is afraid of everybody.

WHAT ELSE COMES TO MIND WHEN YOU THINK OF YOUR FATHER

Bullies.

DOES THAT HAVE ANYTHING TO DO WITH THE FACT THAT YOUR BOYFRIEND MADE YOU COME HERE


Bibliography

Hofstadter, D. (1995) Fluid Concepts and Creative Analogies, Perseus Books.

McGuire, M.T., Lorch, S. and Quarton, G.C. (1967) ‘Man-machine natural language exchanges based on selected features of unrestricted input—II. The use of the time-shared computer as a research tool in studying dyadic communication’, Journal of Psychiatric Research, 5(2), pp. 179–191. 

McCorduck, Pamela (1972) ‘Joseph Weizenbaum -- Transcripts’ Carnegie Mellon University Archives, Box: 3, Folder: 8. Pamela McCorduck Collection, 1978-0001.

Project Mac (1966) Project Mac Progress IV 1966-67,  p. 117-129.

Turkle, S. (1997) Life on the screen: identity in the age of the Internet. New York: Touchstone.

Weil, A. (1965) “Conversations with a Mechanical Psychiatrist.” The Harvard Review, Vol. 3, No. 2, pp. 68–73, 1965.

Weizenbaum, J. (1966) ‘ELIZA - A Computer Program For the Study of Natural Language Communication Between Man And Machine’, Communications of the ACM, 9(1), pp. 36–45.

Weizenbaum, J. (1967) ‘Contextual understanding by computers’, Communications of the ACM, 10(8), pp. 474–480. Available at: https://doi.org/10.1145/363534.363545 

Weizenbaum, J. (1976) Computer power and human reason: from judgment to calculation. San Francisco: Freeman.