I first encountered the mechanism of the meme in Blackmore's The Meme Machine just prior to the Conference on Consciousness (Tucson, April 2000). I had been aware of the concept of memes for many years but had found no truly useful (or rigorous) description. After Blackmore's plenary talk, I saw clearly how the meme was the means to combine the Conference's various explications of consciousness with goals of AI.
One result was my poster session on Digital Immortality at the Conference in 2002. But the more I examined memes the more questions I had. And the more I examined and the more I questioned, the more conjectures I posed. Although obviously intrigued by memes, they are not my primary interestyet I feel compelled to do something with all these questions and conjectures.
First Note to Susan Blackmore
Your book is the book I was looking for when I first became interested in memes some three or four years ago. Eventually, I overcame the negative reviews and read Brodie's book a few months ago. I was pleasantly surprised, but far from satiated.
Through my customary desultory explorations, I noticed your book only a few weeks before the Consciousness conference. I added it to my list of books of interest along with a dozen others I had spied during a Barnes and Noble perusal. I had stumbled across the conference during similar wanderings at Borders only a week before the conference.
I found a copy of your book in the library just a few days before the conference and rushed to devour it before your talk on the last day of the Conference. I was so immersed, that when I encountered you at the poster session Monday night, I said, "You're the reason I can't sleep." You apologized for your jet lag and we both moved on.
Before your talk on Tuesday, I had excerpted the following notes from your book:
"In the brain, every input builds on what went before." (Page 57) A simple statement, but one I firmly believe in. All our thoughts and concepts can only be constructed on what came before. Present understanding is always limited to the ideas, thoughts, and concepts we already havethe tools at hand. The problem is to know when these are inadequate; when to fashion new tools of thought.
On page 127, you referenced evolutionary psychology, placing memes squarely in the one of the main streams of consciousness research. To suggest any mental mechanism as essential, one must show its evolution as necessary for our own.
You state on page 141, "Memetic theory easily explains why the education of women is so important in changing family size." This reminded me of Buckminster Fuller's assertion that the primary solution to the problem of overpopulation was options for women.
"[Genes have an] . . . exquisitely high-fidelity digital copying system based on DNA. The memes have not yet reached such a high-quality system and will probably not settle on one for a long time yet." (Page 209) True, insofar as we are the medium. However, suppose the strategy of memes is to implement some form of intelligence in computers; these not only have capability of perfect copying but they also evolve much faster than we do. We need to be aware of these implications.
On page 213, you stress the power of copying the instructions. This rang a very loud bell for this ex-computer programmer. I wonder if anyone who has not revised code into more powerful abstractions can fully appreciate the power of this statement.
The suggestion is the user illusion of self is derived from copying others, i.e., they appear to have intentionality, so I should too. Regarding the intentional stance, you say on page 229, ". . . we believe 'as if' other people (and sometimes animals, plants, toys, and computers) have intentions, desires, beliefs, and so on." True. In addition, for millions of years we believed the same of sun, moon, stars, the winds, and the weather. Prehistoric humans had the belief (as many people now still do) that animals, and sun, moon, wind, and stars were animated selves (from anima, soul). However, this causes me to wonder why we assume it is our own sense of self that causes us to see a self in others. This seems to ignore our proclivity for copying. Perhaps it's the other way round: "That 'other' seems to have intention, so why shouldn't I?"
Why did we do this, developing self-awareness, and animals didn't? A: Because we evolved better copying skills.
Notes from Plenary talk on Tuesday April 11, 2000
1. The self is an illusion 2. Why? 3. Evolution.
Why should we be surprised to discover that self (the selfplex) has problems? Like other evolved structures don't?
The inevitable result of memes will be AI! My earlier comment about the need to understand Consciousness as prerequisite for AI. This is the overarching meme of this Conference. (So who here is pursing memes toward AI?)
My particular interest in memes is Social Invention.
Is attempting to imitate a skill a meme?
The Selfplex protecting its attacked memes. (An attack on some portion of who I am, is an attack on all of me.) This is highly, highly significant.
Some of us know we are more than our memesbut this must be learned.
We cannot understand Consciousness and Self w/o memes!
Q: What is the memes' contribution to the illusory nature of Consciousness?
Q: Is our difference because animals are limited in their ability to gesture?
What is the extent to which animals "fill-in"?
Animals also differentiate self from environment, so they may have some degree of consciousness. Is, in fact, the major difference simply memes?
Trying to discover how the mind works from within your own head is not unlike the mentally ill attempting to cure themselves. It's all in your head.
We must keep reminding ourselves that the brain is not a passive device.
My follow-up questions and conjectures
Is the meme of immortality inherent in consciousness?
Isn't this what Blackmore is saying? That the self is merely a meme?
Blackmore seems to assume all memes are in favor of promoting memes (in general) over genes. Why? Shouldn't any given meme only be "interested" in its own propagation (which may or may not promote other memes.)
The Meme Machine, pg. 105: fast talkers and sex, i.e., "being highly articulate makes you sexually attractive." But what's the need for indirection? Many highly articulate people use those skills directly for sex. (HG Wells)
Language & Memes
The key element in the origin (or evolution) of speech is imitation, the transmission of information as memes. But did this truly begin with language? Is anyone looking into Gestures as Memes, i.e., Pre-linguistic Memes. Also, what is the relationship of sign language to gesture? What is Blackmore's Rush to Language?
- Language is a subset of communication
- Gestures precede language. It's what we all do when there's no spoken language in common. Long before there was: "Me, Tarzan; you, Jane" we pointed and poked, waved for others to come or go, and mimed actions past or threatened in the future.
- Memes can also be (any form of) communication.
- Communication evolved for basic needs: sex, food, power (Brodie). All the higher animals clearly communicate about these.
- Humans, who were better at copying, communicate better by copying memes.
Some Universal Gestures: Hand to ear indicating not hearing properly; Shrugging shoulders for don't understand; Most common group: pointing, motioning to come this way, stop (pushing away). "They went that'a way." The predominance of gesture in pre-lingual man.
Is the smile universal? Not a gesture, it still signals information. Others?
It would seem that somewhere there must be a glossary of universal gestures. What's important is showing how our gestures differ from the animals (some are same, e.g. baring teeth); how they differ.
Sounds such as laughing and singing obviously predate words.
If anything, gestures are the intermediate step to language. Gestures could become memes, embody meaning, denoting objects and actions. Then, sounds (which were also part of the pre-human environment) become associated with gestures and words begin to develop.
Should we not then look for the earliest gestures? For example, the (pre-historic) father draws back in hand as if to hit the child but doesn't. I.e., "I could hit you but I won'tfor now." The child then copies and demonstrates his power over a younger sibling and we have a replicated meme. (And Ralph Kramden gesturing as he says, "To the moon, Alice, to the moon.")
Advantages of Language over Gesture: dark, distance, close by (whisper).
Communication is born of imitation through the meme! (But not necessarily via language first.)
Preparation for Introduction to Memetics Workshop for 2002 Conference:
Process, Objects, and Dualism
Just reading Blackmore wherein she says the mind stuff of dualism does not exist because we cannot "locate it". (Penrose's and Hammeroff's microtubules are so much fairy dust.) But location means place, and it presumes object. For example, we use these computers (like the one I'm writing this on), but few people could locate the program (analogous to mind) that controls the hardware (analogous to body). Far fewer still could find that part of the program controlling the input, correction, and display of these words.
Yet, these exist. Someone wrote the code that performs these functions. Someone else wrote the operating system code that controls the operation of this word processing program. As written, these programs exist in human readable form, a printout of all the instructions. But that printout, that object, is not what is performing those functions as I type. It is the functioning code in the machine that makes it difficult to locate where which component does what.
To narrow it down: What I interact with as I type is the machine's execution of the program's instructions. This is clearly better described as process rather than object. Processes are much harder to locate, and much harder to think about, than objects. The process of your aging is magnitudes more complexand ephemeralthan the state of your body, an object, at any given moment.
Our primitive (in an evolutionary perspective) brains are much better at thinking about objects than about processes. In fact, when we encounter processes we don't understand, we think of them as attributes of invisible objects. (For example, the ever popular volcano god.) These unencounterable objects can be alive, yet they cannot die. They can have power over our lives, yet have no physical contact with us.
Process can be described as a series of state changes. States of what? To simplify, we think of them as states of objects. But if we look deeper, we must acknowledge these, too, are not objects but also processes.
Photosynthesis is a process we understand fairly well. It involves many such processes (parts and whole) and many changes in their states. Human vision (not to mention consciousness) is a process and sub-processes we understand less well. One thing we do know about vision is that the eye is not like a camera. For one thing, cameras are passive and vision is an active process. Seeing is not about recording, but understanding. So, too, is our consciousness activeattempting to apprehend, not simply record.
Notes from second reading of The Meme Machine
"The inevitability of evolution is part of what makes Darwin's insight so clever." (page 11) And so dangerous, in that it clearly precludes the need for a prime mover.
"Memes have become the tools with which we think." (page 15) It seems to me that the intelligent and educated overvalue their thought processes. That somehow their thoughts have unlimited degrees of freedom, when in fact they, like all thinking creatures, are completely constrained by their current bag of tools of thought. Fortunately, we are capable of progress because we can improve and refine these tools and acquire new ones. But total freedom of thought? Nonsense.
"Human consciousness itself is a product of memes." (page 22) While this might seem the key proposition for memetics, it does not mean the theory cannot contribute to an understanding of consciousness if this proposition is only partly true.
"Conflicts of interests (between genes and memes)." (page 30) Seemingly obvious, but how can we make use of this knowledge?
"They [infants] seem to know when they are being imitated by adults." (page 50) Having directly experienced this gives a whole new appreciation for their skills.
"By making discrete words instead of a continuum of sound, copying becomes more accurate." (page 102) The movement from analog to digital enhances information flow and reproduction.
"Copybots." (page 106) Sounds like a clearly defined stage en route to AI.
"We would expect cutters with the least horizontal transmission to have the strongest taboos and vice versa." (page 137) I had an earlier note to find a way to differentiate memes that easily replicate throughout society from those we may personally acquire (a parent saying, "You are stupid.") but are unlikely to replicate.
"Memes are busy devising ways of interfering even more directly with genes." (page 145) Again, similarly to their conflict of interests, how do we make use of this knowledge?
"They [memes] are still evolving their copying machines for and this is what all the technology is for." (page 204) While we are good meme replicators, clearly technology has the capability to be even better. What then of us?
"For robots to become like humansin other words, to have human-like artificial intelligence and artificial consciousnessthey would need to have memes." (page 218) Exactly! And why not? And why not see this as the method to build AI, i.e., first create meme machines and then look for intelligence and consciousness?
"The harm that comes from having a false sense of self." (page 246) Or simply an inaccurate one. Brings a whole new meaning to "Know Thyself" doesn't it? Of course, first we must have a better understanding of just what a self is. For example, in multiple personalities it seems only one self at a time can have consciousness. Therefore, it would seem that the meaning of self and the meaning of consciousness are inextricably intertwined.
Notes to S.B. from Workshop on Sunday April 7, 2002
You mentioned Urban Myths as examples of memes. However, you neglected to mention the counter-trend, i.e., there are thousands of websites debunking Urban Myths. Is there a need for a category of contra-memes?
In discussing Darwin, you aptly described evolution as "design by extinction." You also correctly emphasized that "Inevitable is key word in Darwin."
I found it very interesting when you stated there were no creationists in the UK. Amazing, given this was the home of the original anti-Darwin movements (Bishop Wilberforce and all that).
Since I have been writing about this particular problem for decades, I appreciated your example of short-term capitalism as a successful long-term dysfunctional meme.
When you pointed out that memes required fidelity, longevity, and fecundity, I realized Digital Immortality replications would not have fecundity! Which caused me to wonder if copying was an adequate substitution?
Another cause for wonder is, what does it mean to say "we" choose memes? If "we" are no more than our selfplexes, then don't those memes (that collection, system, whatever of memes) do the choosing?
Curse words as very effective self-contained memes. Also jokes. But how and when do we differentiate between those memes which are self-contained and those which are not? And do those that are not require further subdivision?
Mediums for memesneed better word. Not instruments. How about channels?
Proto-Languages: were there any? What were they like? Why do we never see this? I.e., do they develop so fast we never see this stage?
Location of memes: Objects vs. Process. See previous discussion on Process, Objects, and Dualism.
We have got memesor do they have us?
Copying through transformation becomes symbolic! Wish I knew what I meant by this. Certainly seems like it needs more explanation. Related to W. Percy?
Copybots as experiments. On web as Talking Heads (Steels). Need to look this up. Also emergent grammar (proto-language?)
Mentioned co-evolution. This is reminiscent of G. Bateson. Any connection?
Consciousness/Continuity/Self. Are memes necessary? She thinks not! In what way? That we could have some form of consciousness (undifferentiated aesthetic continuum) without memes is true. But would we have continuity without the fiction of the narrative creating and maintaining self?
Memes are a Bundle Theory (Hume). My note says only she and Dennett believe in this. True? E.g., her comment?
Persons (according to Dennett) have rights. And digital persons?
Isn't Dennett's Multiple Drafts Theory from Calvin?
What does it mean when people think and talk of themselves in the third person?
She said she didn't know how to label consciousness empty of memes. Wouldn't that be the moment of the undifferentiated aesthetic continuum? (from F.S.C. Northrup)
Follow-up questions and conjectures
Do I recall correctly that she referred to Consciousness as the Persistence of Memes? How about "Persistence of Memory" by Dali?
On this planet, genes evolve through both human and non-human media. The same is true with memes! (Especially vis the conflict between humans and technology) The question for the future is, while we may at present be part of the process in the evolution of non-human memes, are we at all necessary?
More than simpleand immediatecopying, it is the remembering of the meme (words, gestures) and its reasonably correct application is appropriate future circumstances. In other words, copying is more about the copying now in the mind for recall and use later. E.g., her example at workshop was first you, then you. So are we then back to the importance of memory plus meaning? (Memes requiring both.) 1. Memory needs to be sufficiently accurate and retrievable. 2. The Longevity Factor (Persistence).
Horizontal spreading of memes succeeds due to numbers rather then strength of transmission. Whereas vertical transmission has far fewer numbers but succeeds, when it does, through the strength of the transmission.
Does this mean we should do more definition of memes as regards horizontal (more cultural and social) versus vertical (more personal and behavioral)? For example, the meme, "You are stupid" is not likely to become passed on horizontally to the rest of society, although it may be passed vertically within a family. Can we do more to distinguish these two forms? And if we do, how does this affect our "treatment" of the two forms? E.g., the memes accompanying child abuse (vertical) versus politician's slogans.
What about the use of memes in the replication of behavior, i.e., acting!
More needs to be done with the problems of the selfplex as a consistent system. For example, we hear prostitutes say they separate their commercial activities from private life, i.e., "It's not me doing those things." For another, people who lose close life-partners, frequently speak and behave as though they've lost a significant portion of self!
Capitalism vs. Communism: Is a meme which is reinforced by positive results (20th Century Capitalism) a stronger replicator than one which promises everything (Communism, Religion) but can't be tested for results? To restate: Capitalism was a more successful meme because of its results, whereas Communism failed because it only promised.
More thoughts on Copying
We are the Replicators (Copy Cats).
Imitation and peer groups.
Imitation as meme replication.
The process that creates a mental construct is imitationthe meme process is the essential tool of thought.
Looked at in this way, what we call "filling in" (Dennett not withstanding) is also imitation. I.e., imitating something we know (or see) in assessing knowledge (or vision) to "fill in" where it is lacking. Therefore, the process of imitating (meme) is essential to all mental functions, not just thought. Is it then possible that the basis for the capability of memes resides in the automatic imitation that completes (visual or mental) pictures by copying surrounding areas into gaps?
Could rehearsals in mind be seen as imitations (sometimes scenarios) of future eventsand therefore memes!
Is it possible to sum this all up as Memes for Dummies? E.g., begin with gestures, sounds; differences with animals; all the way to how memes are moving to AI.
Note the base of all these is "logical." We are aware of genes (biological) and memes (bi-logical), but need to add technological. (Does McLuhan talk of extensions as replicating?) Important point: technological replication comes from an interaction (combination?) of genes and memes! How about other logicals? E.g. Chem-, Phys-, and . . . ? (Note that the path of digital is a long rising arch whereas biological is that curve of S. J. Gould's with a peak close to the left vertical and a long trailing off to the right.)
- Technological (e.g. McLuhan) also history of technology
- Socio-logical (memes) Although this might be seen as predating technology, it's only recently been defined as a replicator. (Is the point here that tools predate language?)
- Bi-logical: the binary logic of the computerthe future of memes.
This is a story about replicationand where it is taking us. First, it was biological, and messy. Then there was technology. Fire preceded the wheel, but inventions like bags and pots must have been even earlier, e.g. carrying food and water; then clothing for colder climes. Now we're entering the digital age where perfect reproduction is possible. And what it all means.
Until this last century, memes could only replicate through imperfect evolutionary biological media. Changewhich we see as healthy in living things replicated by geneswas inevitable, part of the process. No longer. While most memes are still subject to the same evolutionary forces, a few have achieved perfect reproduction thanks to the rise of digital media.
The question (certainly the question for AI) is whether these digital media can also be evolutionary. Genetic algorithms would seem a possibility, but then the question becomes when to apply which of these forms?
Printing, recording, and video all precede digital, and all are pretty good replicators; i.e., far better than handmade replication. (Clearly, the rise of machine replication leads to the best machine; that is, digital.) Take examples of video blurring after several generations, old movie prints, and books missing pages. But none of these is a problem for digital.
However, despite the superior replication of digital for better meme reproduction, we can't yet use it for AI. This is the fundamental problem. We can't design/create any viable AI without knowing how our own intelligence (and by inference, other forms of intelligence) works. We are beginning to understand perceptual systems for example, but as long as we think of mind as an object (mind as separate from, or somehow integrated with, body) we are still in the Stone Age.
We are only dimly aware of how we acquire information (from sense data) and how it becomes knowledge and sometimes wisdom. And while we are beginning to know something of genes, we are barely (and rarely) aware of memes. E.g., if evolution is gene-directed and intelligence is meme-directed, then do we really know enough to begin designing/creating vessels for either?
And despite any observations that memes are pushing towards AI, we still need examples, specifics, demonstrations, experiments, etc. And a clear exposition of whyfrom the meme's point of viewthey are doing this.
And what of emotions? Gelernter says we can have emotional machines and I believe Blackmore also said something about emotions are needed for anything artificial to be human-like.
If I were the (large) memeplex known as Plato's Republic, I would have been thrilled when I was first reproduced on a printing press. I would have then become ecstatic when I was uploaded onto the Internet. I will be further joyed when my first copy is downloaded into a handheld electronic book (or that "intelligent" paper Negroponte speaks of).
The advantages of digital are obvious for reproduction, so why would this memeplex not support further advances in this direction. But, and here is an interesting but . . . . does this replication enhance Plato's Republic's competition with other memeplexes? (As being read and argued by humans.) Is this no longer necessary, with digital space large enough for all memes? In other words, is replication without competition all any meme could want?
Here we come up against the question of whether the digital replication of memes will produce the same evolution memes have enjoyed to date. Specifically, if memes require fidelity, longevity, and fecundity, then doesn't digital reproduction exclude fecundity? Most certainly, digital is far superior in both fidelity and longevity, but where are the variations needed for true fecundity? Certainly, perfect reproduction would not have lead to our current evolutionary chain. Or will digital memes find another avenue to fecundity?
Also: How does this relate to Calvin's six essential aspects of the creative Darwinian process that bootstraps quality?
- reasonably complex pattern
- pattern copied somehow
- variant patterns by chance
- pattern and variant must compete (for limited resources)
- multi-faceted environment
- successful patterns (survival, sexual selection).
Memes are responsible for the "irrational" choices of people who desire to be guided by reason. Most people here at the Conference think their actions are directed by their reason. But despite their belief in their own rationality, isn't it possible that memes are controlling much of their behavior?
Suicide is the selfplex's extreme attempt to prove itself right.
Watching the mentally ill trying to make sense of their world. How different from us are they really? Clearly, the memes are in charge. This being humanall these variationsare part of the continuum.
Need much, much more on the ability of the selfplex to compartmentalize.
Sternberg (using the verb, elaborate) says our (love) stories become more complex, more complicated, over time. Another way of saying that over time the selfplex becomes more complex. Saying people grow more complex, and therefore more difficult, over time. Typically, many of them are characterized as old and cranky. A good deal of what we call getting old (in a non-physical, behavioral sense) is exactly the difficulty of dealing with this growing complexity of the selfplex.
If you examine people who appear younger than their chronological age, you find people with less complex selfplexes. (For example, the extreme of people in their "second childhood.") As we age, we need to recognize this growing complication of our selfplex (of our selves). And to counteract it by learning (as Thoreau among others suggested) to simplify. When the selfplex becomes too complex for its own good, we need to shed, to simplify.
Blackmore referred to memes (if I recall correctly) as time-binding. I wonder if she is familiar with Korzybski.
We accept, even if we don't understand, when the criminal says, "I couldn't help it." We accept this statement as evidence of his or her inherent criminal mind. Don't we also see ourselves as also having an "inherent" nature we can't change? When people say they can't change a behavior because it's who they are, they're talking directly about their memes! Knowing this, how can we use it to help people make changes?
Need to know more detail of the mechanism of meme replication through mass media.
Need to know more detail of the mechanism of meme replication through mass media.
Tools are most certainly memes.
And evolution of memes? Our idea of medicine has drastically changed in the past 500 years. And significantly in past 50. Again, how to acquire the detail of the process?
At the end of the Conference, I gave Susan Blackmore this continuum: Genes/Species/Society/Tribes/Family/Individuals/Memes. The idea was that genes affect memes by first affecting the species, then society, etc. And, conversely, that memes affect genes by first affecting individuals, then families, etc.