Electric Dreams

The DreamSpinner Column: 

Working Dreams With The Power Of Computers

Bjo Ashwill

(Electric Dreams)  (Article Index)  (Search for Topic)  (View Article Options)

  Ashwill, Bjo (2000 July). The DreamSpinner Column: Working Dreams With The Power Of Computers. Electric Dreams 7(7). Retrieved July 14, 2000 from Electric Dreams on the World Wide Web: http://www.dreamgate.com/electric-dreams

Hi, Electric Dreamers. I am Bjo Ashwill and am writing a monthly column on my experiences of creating a computer software program that does very detailed analysis of dream narratives. You are welcome to visit my web site and check out DreamSpinner, the software program I will be describing. http://www.spinner-soft.com.

In this column I shall describe, over time, how to use the computer's power to store, group, analyze and retrieve information from our dreams. DreamSpinner's greatest power is working with long "over time" dream series, although it can work with individual dreams as well. How do metaphor patterns change over time? That is the question that began my journey toward creating DreamSpinner.

Following is an abstract I wrote for the ASD Conference in 1997 describing how I created categories in DreamSpinner.



"Written Dream Narrative, and conceptualizing a categorization system for a computer."
Presented at ASD Conference in 1997

Dreams are recorded in many different ways. Paintings, dancing, collages, poetry and of course the dream journal where a dream narrative is our attempt to translate our wildly vivid and densely packed dreams into words. Dreams translated into Dream Narratives is literature, or at the very least, written language. As such, it conforms to the useful conventions and concepts used in linguistics when studying a body of text.

Language is fluid, flexible and constantly changing. How do you pin those words (lexical units) down? It's a bit like the chaos theory. The more you look, the more there is to see. Now that we have computers with speed, power and large storage capacities, we have the tools needed to explore dream narratives from many different angles, which we could not do with hand written journals or series.

But how do you categorize thousands of words. There are of course as many different ways as there are dreamers wanting to categorize. Here is how I chose to do it for DreamSpinner.

Major premises I had when creating DreamSpinner's categories:
1. Keep the category system as flexible and open ended possible. People are constantly making up new words and different meanings for words all ready existing. The language is a dynamic fluid thing. There has to be room for that fluidity in the categories.

2. The person doing the categorizing (DreamLinking) is often the dreamer themselves. So coder consistency isn't as important.

3. Finding consistencies, patterns and repetitions of motifs is a central task for DreamSpinner.

4. All words have relevance and information to reveal.

5. The Dream Narrative is a linguistic text.

6. You can link a word to more than one category.

I began by "melting down" my 3,000 dreams into an alphabetical list of all the words I used. I assumed that I would have a sufficient enough vocabulary to encompass the most frequently used words. Turns out that is correct. Statistical prediction indicates the first 15 words on a frequency count list will account for 25 percent of the text sample. The first one hundred words will account for 60 percent; and the first one thousand is used in 85 percent of the sample. Checking the most frequently used words is like taking a blood test. Is the white blood cell count up or down from the norm? This is valuable information. However, there are many one time usage words that are individualistically unique. Psycholinguistics and Stylolinguistics learn much through studying the one time usage words.

So I now had a list of 10,000 words.

I had to determine what is a word. Is it the root word, or is each syntax variation of a word a separate word. Is a different meaning of the same word a separate word. Is a phrase a "word"? I decided to work with the material as root words, incorporating the syntax variations under the one root word. If the word had more than one meaning, as most did, then I would give a separate root word connection for each different meaning. I used the Parts of Speech categories, keeping verbs, nouns, adjectives and adverbs separated.

Hit = Take a toke off a marijuana cigarette. (Category: Street drugs, sub category of medicine.)

Hit = Physically strike someone with your body.(Category: Motion and if aggressive, Category: Social Interaction, then sub-category Aggressive.)

Hit = Strike a baseball with a bat. (Categories: Games, Motion, Recreation Related)

Hit = A popular song at the "top" of the charts. (Category: Music, Up)

(There are more, of course.)

Now I wanted to hook each root word to at least one category. DreamSpinner itself was born of ideas inspired by Calvin S Hall's "The Meaning of Dreams". I found Bill Domhoff's and Adam Schneider's Quantitative Analysis of Dreams Web page and immediately began trying to match my word list from my dreams into the categories the Hall/Van de Castle coding system had spent years developing. I certainly couldn't come up with anything better.

I ran into a problem though. A great many of my words simply did not fit under the Hall/Van de Castle categories. I was using every word in my dream narrative. So I had to start inventing, melding, and stretching categories beyond what the Hall/Van de Castle system indicated. Many of the Hall/Van de Castle categories focused on concrete words, like names of objects and action verbs. Many of my words were abstract, prepositional or connecting in nature. I researched linguistic materials for category ideas and managed to get all my root words under some category or other.

By this time DreamSpinner had progressed to the place where I could begin linking each word in the dream to the specific root word which is then hooked to one or more categories. Some categories weren't working. Many words had essentially the same meaning whether used as a verb, noun or adjective. So we elected to toss out the differentiation in Parts of Speech. Categories became morphed, renamed, cut up into pieces, or blended together until the mix met the actual data experience of my dreams.

In order to create the category system I now have, I researched:
1. Parts of speech.
2. Scoped out dream books for dream themes.
3. Checked out thesaurus for basic headings.
4. Looked at psychological theory for important areas of concern for humans.
5. Examined already existing categories, both in the dream and linguistic worlds
6. Used my dreams as the raw data to measure the categories with. If categories remained empty, they were removed, if seemed to general, they were split out into sub-categories.
7. Created categories like "Problem Solving Techniques", "Object Relationships", "the Shadow" and "Nuances", with sub-categories like "Male/Female Blending", "Virgin/Whore Blending" or "Incongruities". These goes beyond the dictionary definition of a lexical unit and puts together words not necessarily related to each other in meaning, but related to a common theme. Chasing the possible metaphors. For example, a pipe would be in the category of a building detail , under Architecture. Also under Flow, because it is a conduit for material to move through. And under body metaphors.

This brings us to the end of the Abstract reprint. Following the references, is the next step. Metaphors!! Are they Universal?


Domhoff, G William, Finding Meaning In Dreams: A Quantitative Approach. Plenum Press New York 1996.

Domhoff, G. William and Schneider, Adam. Quantitative Analysis of Dreams Web Page: http://zzyx.ucsc.edu/~dreams/Coding_Rules/

Goldberg, Lewis R. (1990). An Alternative "Description of Personality": The Big-Five Factor Structure. Journal of Personality and Social Psychology., 6, (pp.1216-1229)

Hall, Calvin S., The Meaning Of Dreams. McGraw-Hill Book Company, New York. 1966.

Huang, Shuan-Fan. A study of adverbs. The Hague : Mouton, 1975

Kucera, Henry and Francis, W. Nelson, Computational Analysis of Present-Day American English. Brown University Press, Providence, RI. 1967.

Lakoff, George and Johnson, Mark. Metaphors We Live By University of Chicago, Chicago, 1980.

Levin, Beth . English Verb Classes and Alternations: A Preliminary Investigation. University of Chicago, 1993.


Using the process I described above in the abstract, I created a giant dictionary which is organized in a thesaurus like category system. I developed the capability of not just searching single words (as metaphors) but related words together. Dream Dictionaries we are used to seeing give a single word and try to make a general statement about the possible meaning of that word. I have expanded the concept of metaphor because I believe that metaphor is our mother tongue. That's how we think (check out the reference to George Lakoff's books.) Therefore, my philosophy is that ALL words used in a Dream narrative are metaphors.

Water, for example is considered a universal metaphor. In my view, it is not enough to examine the metaphor "water". The water is a specific type of water (lake, ocean, puddle, drinking water, mud, tears) and the actions and feelings about the water is critical to the possible personal interpretation of what water means in MY dream. Is it tidal waves or calm water? What colors are involved. What happens on or in or near the water? DreamSpinner, because of the unique category system and the linking process can pull up a dream set where all dreams have lakes in them, and only those lake dreams with agitated water is called up. The frequency count that comes up for that dream set will tell me hundreds of fascinating facts. Who is most frequently found in those dreams? Are there more or less animals involved in those dreams? What is the most frequent emotion? Armed with this information, I can formulate what the metaphor water (in relation to lakes and agitated water) means specifically for me.

Doing my own dreams, I learned that most of my lake dreams had something sinister and scary lurking under the surface of the lake. There is a high incidence of strangers. The emotions are most commonly terror and fear. What a gift that is to see that recurring pattern that I wasn't even looking for.

Now I want to be clear about my bias about metaphors NOT being universal except in a very general sense. Because MY lake dreams indicate that pattern, it does not mean YOUR lake dreams mean that at all. Only after collecting dreams from many people and DreamLinking them will I have the ability to compare and contrast and make any formal statement about "Lake" dreams, or any other metaphors I examine.

My Website, Http://www.spinner-soft.com, is in the middle of a total reconstruction. When the new version is up and running (soon, I hope, but remember "Soon" in computer terms is a very flexible term. "Soon" could be weeks to a few months.) then I will have a call for dreams to use in the DreamLinking process to create those norming dream sets. You will, if you wish, be able to enter your dreams and view dreams others have entered. You can choose to allow your dream to be used in the norming process or not. It is entirely up to you. In addition, I will be creating a Metaphor section where we can share dreams related to specific metaphors (categories). We will then see, right before our very eyes the unfolding of the answer to the question "Are Metaphors Universal or personal, or both!!"

Next Month, I will be more specific about the category (metaphor) system in DreamSpinner and use some Dream Set examples. If you have a specific metaphor you would like to be addressed in next month's column, or up on the Website, please email me. Dreambjo@hotmail.com. I wish you powerful, insightful dreams!