Thursday, July 10, 2008

Memory, Memory

Excuse me for starting a new post rather than appending to Howard's previous post. After thirteen comments it seems more convenient and better focused.

Howard's post calls dna a kind of memory. I agree that dna could be thought of as a sort of "species memory" for certain purposes, but I don't like this in a more general sense when one looks at the time factor. A current strand of dna is the result of multiple copying errors and mutations that have occurred in the past. Just because something is a RESULT of occurrences in the past doesn't mean that it is a RECORD of that past. I would prefer another approach based upon speculation about evolution.

The closest thing that I can think of to memory at a cellular level is "tropism," which may be the ancestor of true memory. For example single cells may be attracted to sunlight and away from shadow. They are wired to do this based upon natural selection and the nature of their food source. This sort of "pre-memory" or "proto-memory" records the fact that the more food is and has been associated with the light. It isn't true memory, because although it has both permanence and changeability, it isn't addressable. It's more like the set point on a thermostat. Although circumstances can cause a change in the setpoint, the current setting tells us nothing about the history of thermostat's settings.

To better understand the evolution of memory we need to ask why memory might be of value. In the case of photo tropism I'll create an imaginary situation. Suppose that the food versus light intensity relation was also a function of light frequency. A cell might evolve which had different receptors for different frequencies. Each of these receptors would have a different intensity setpoint. Hence, we have stored the information of the best intensity for the cell to propel itself toward, depending on the frequency of the light. The optimal intensity is therefore "addressable" in the sense that only that sensor is activated which corresponds to the particular frequency in the environment. A mechanical analog would be a combination thermostat and humidistat. Suppose we prefer various temperatures in a room depending on the humidity. Imagine that the humidistat selects a different thermostat with a different temperature setpoint depending upon the value of the humidity. We now have true memory. The array of thermostats contains all the temperatures existant and the humidistat makes them addressable according to the humidity. In effect saying, "I remember that when the humidity is x% you like the temperature to be y degrees." If there is any analog to this at the dna level, I believe it would be at the level of expression. A change in environmental conditions in the womb may cause the expression of certain genes to be encouraged or discouraged. Hence, the very same set of genes may produce different results, because of the inhibitors or promoters attached to the dna strand. Those inhibitor and promoter particles (I've forgotten their names) might be thought of as memory.

To summarize might we say that the existence of memory requires that an action shall be taken or not taken, depending upon a stored variable which is addressed by an environmental variable. At a macro human level this would be that a piece of information comes to consciousness from among many possibilities depending upon whether or not one finds the right access path or phrases the right question.

With respect -Joel

27 comments:

Ira Glickstein said...

Well, after a dry period of some weeks, we have two new Topics posted on the same day!

Thanks Joel for digging more deeply into Howard's analogy between human (or other animal) memory and DNA memory.

As you point out, there is a wide difference in the time factor and what you call being "addressable".

In the case of human (or animal) memory, it is accessible in "real-time". In the case of what Howard and I call DNA memory, it takes generations to access the memory.

Your example of a system with memory setting a different temperature according to humidity is similar to the historic case of the "pepper moth".

This moth is white with black spots. Specimens collected prior to the industrial revolution are mostly white with a few small spots. Later specimens have more spots and, from a distance, appear sooty gray. More recent specimens, collected after sooty emissions were reduced due to stricter environmental measures, are once again nearly white!

This appears to be a case of gene expression regulation (This link compares chimps and our President.)

The genes for spots existed and those pepper moths that happened to have gene regulators for more spots were darker and more likely to survive and reproduce and pass these genes and regulators on during the sooty times. Later, with a cleaner environment, the moths with different regulator genes were whiter and so survived and reproduced.

Ira Glickstein

Howard Pattee said...

Joel and Ira,
All I can say is that I prefer the established usage of words, although I would agree definitions are largely arbitrary. The basic necessity is that we agree on the usage.

The term “genetic memory” has been standard since before the discovery of the genetic code. Nirenberg, who discovered the code, began his Nobel lecture with the words, “Genetic memory resides in specific molecules of nucleic acid.”

Of course, there are many different kinds of memory. Memory is briefly defined only by its function of storing information for future use. Both human writing and DNA can be accessed in minutes or after thousands of years.

Howard

joel said...

Memory all by itself is not very interesting. It is only a recording. What is interesting, is what we do with it. The first is the use of ancient or genetic memory to recognize food or danger. An example would be the attraction of bees toward the bright colors of flowers or the complex muscular action of suckling in mammels. The next level might be the use of learned memory in an individual to learn how to function in the present. For example creature as tiny as fruit flies use memory to record patterns for the purpose of navigation. Visual pattern matching is not the only kind. Patterns of sounds and smells are also matched in nature in order to orient (in the most general sense of the word). We might categorize this as using the past to understand the present. In man we find a highly developed ability to use the past to predict the future. That isn't to say that that this capability is unique to man. We need to consider the extrapolation of prey trajectory to be a simple type of prediction of future from the near term past. Clearly, there are many creatures with this talent. Although man thinks he has a much enhanced ability to plan based upon memory of the past, it's not clear whether much of this ability is imagined rather than real. Experiments with lower primates indicate that they are capable of planning or problem solving also.

Aristotle thought that it was important to determine the essential nature of a creature in part to determine what was worth enhancing. For example, the essence of a horse is speed. Thus a race horse is the epitome of "horseness." It is worth seeking the essential nature of a human in order to determine what a human should enhance (again, according to Aristotle). In the above discussion of the usage of memory we see that much of the animal kingdom has the capacity for genetic memory templating, experiential memory pattern recognition, self orientation and extrapolative planning. One might say that the ability to use memory for planning is the essence of man, since his ability to plan, though not exclusive, far exceeds that of other creatures. I propose that there my be an application of memory that is absolutely exclusive to man.

It may be that only man uses memory to reminisce. That is to say recollection purely for the sake of recollection. Surely this sentimental journey of the mind has no survival value in genetic evolution and is not evident in the rest of the animal kingdom. It's ironic that we come full circle. The roots of the word memory lie in the Greek concept of poetic memory and the ability to perform reminiscence before an audience.

With respect -Joel

Ira Glickstein said...

I was going to challenge Joel's contention that "It may be that only man uses memory to reminisce. That is to say recollection purely for the sake of recollection." However, he may be correct!

I was going to argue that my dog, when a puppy, had a cat claw his nose and, for the rest of his life he gave cats a wide berth. Did he remember the details of that particular clawing or did he simply record a general rule in his memory about giving cats a wide berth? Clearly, he generalized that memory to all cats, not just the one who did the deed.

I once read that some psychologist used his young son for experiments, in one case giving him a fur pelt and then scaring him with, I believe, a loud noise. Thereafter, his son reacted to any furry animal with panic. Did the son remember the particular fur pelt and the loud noise or did he simply generalized that memory to all furry things?

Of course, when we use the word "reminisce" we usually imply some fond longing for the past, i.e., "nostalgia" from the Greek νόστος nostos "returning home", and άλγος algos "pain". Funny how our modern use of the term generally omits the "pain" part of remembering the past!

In any case, after thinking about it, I agree with Joel that humans are probably the only animals that really reminisce. Of all animals, we are the ones with the longest time horizons, both past and future! Our memories of the past, both those we replay fondly and those we find painful, are useful in predicting and planning for the future. For that reason, I take issue with Joel's further assertion: "Surely this sentimental journey of the mind has no survival value in genetic evolution and is not evident in the rest of the animal kingdom." Our long, bi-directional time horizon probably makes us better at predicting the far future than any other animals and that probably has a positive affect on our genetic fitness.

Ira Glickstein

Howard Pattee said...

I have been fascinated by spider communication for many years ever since I discovered that jumping spiders paid attention to my face (See Wiki “jumping spider”).

It is incredible how such a small brain (maybe a millionth the volume of human brains) communicates with sight, sound, and smell, hunts, fights, and mates with complex strategies, and constructs and repairs many types of webs. See these two sites, for example.

http://www.biology.uc.edu/faculty/uetz/vc.htm
http://jeb.biologists.org/cgi/content/full/206/22/4029

I don’t want to argue that spiders think. On the other hand, thinking is not empirically defined, and in any case, thinking must have evolved gradually, so we might learn something about it from its precursors.

Spiders have survived for 400,000,000 years. We’re not likely to last that long, and none of our artificial robotic systems have even come close to the spider brain and sensorimotor system.

Howard

Ira Glickstein said...

I have a vague memory of you (Howard) some years back saying the spider brain had an incredibly small number of neurons - the number I remember is 1200, but I may be confabulating.

In your most recent posting you say the spider's brain is a millionth of a humans, so if we have 10 billion neurons they would have about 10,000, which still sems small for those behaviors described in your linked websites.

Do you know the correct estimated number?

Ira Glickstein

Howard Pattee said...

Ira,

I can’t find any empirical data on the number of neurons in spiders or insects. I assume spider brain sizes are comparable to insect brains, but I have no data on this either. My factor was based on ballpark brain weights (man ~1 kilo; insect ~0.001 gm) and linear brain size (man ~15 cm; insect ~0.07 cm). Insects and spiders also vary in size by at least a factor of 10.

Insect flight is another wonder. Wiki has a short summary of their aerodynamics, but doesn’t mention the visual control system that allows mayflies to remain stationary in a gusty wind and flies to often evade a flyswatter.

Howard

Howard Pattee said...

I want to get back to Joel’s opinion about human memory. There is vehement controversy among the experts on how much of language use is adaptive and how much is just entertaining. The controversy is vehement just because there is convincing evidence both ways.

Joel said: “It may be that only man uses memory to reminisce. That is to say recollection purely for the sake of recollection. Surely this sentimental journey of the mind has no survival value in genetic evolution and is not evident in the rest of the animal kingdom.”

We all know that reminiscing can be enjoyable (I’m enjoying it more and more in my old age), but I still think fictional story telling is the first exclusive human function of language, and it is still a dominant function. The historical records begin with fictional stories that we now classify as myths, epics, sagas, etc. that I believe have survived because they are entertaining, not because they are memories of real events. Furthermore, children’s first use of language, besides complaining, is to invent stories that are often wishful thinking or attempts to get their own way.


The controversial question is: How adaptive is fictional story telling? I tend to agree with Joel, at least to the extent that I see no obvious adaptive value in most stories. However, the ability to devise fictional stories, that is, the creative ability to imagine events that never happened or that one has never experienced,i.e., not in memory, is an essential requirement for inventing those few predictive models that do correspond to reality.

For example, one can view much of science, especially physics, as beginning with imaginative creations like phlogiston, caloric, ether, neutrinos, anti-matter, and Higgs particles, some of which survive in our models and some of which do not.

In other words, we can adapt to our environment because we can create stories, some of which correspond to reality; but that ability necessarily allows us to create many other stories that are merely entertaining.

Ira Glickstein said...

Joel's topic and Howards comments have been quite MEMORABLE!

In this posting, I'll comment on some aspects of memory related to extinction of species, including the coming extinction of the human species.

Let us accept the generic definition of "memory" as anything that is recorded in some media (DNA, brain, paper, tape, ...) with more or less "permanence" and "changeability", and that can later be "addresssed" (or "accessed") and utilized for some purpose.

We each inherit DNA from our biological parents. Our unique DNA is stored in each of our cells. DNA codes for proteins. During gestation and growth, DNA-coded complexes of proteins (which we call genes) are accessed and determine certain genetically inherited traits.

Some of our genes may be unique, the result of random mutations or cross-over between the genes of our parents. However, most go back many generations and many go back 5 million or more years to the last common ancestor (LCA) of humans and chimps.

Point #1: Although the LCA has been extinct for millions of years (if one came back to life he or she could not successfully breed with anyone alive today) many (perhaps most) of our modern human genes are more or less identical to those the LCA modified during its time on Earth and passed on to us. Thus, even if humans go extinct and are succeeded by cyborgs, if our successors have many of the same genes we modified and passed on, we will not "really" be extinct. Our MEMORY will still be in the cyborg genome.

Similarly, we inherit memes from those who raise us (generally our parents, sibs, grandparents, neighbors, teachers, religious leaders, ...) As we live our lives, some of those memes are modified and augmented by those we work with, the broadcast media, the Internet, our children, etc. In addition, some of us may come up with a new meme and add it to the human memenome.

Perhaps some genius among the LCA was the primate who invented the grass blade method of harvesting ants or termites. He or she accidentally or purposely pushed a blade of grass through a hole in an anthill and, when it was removed, it was covered in tasty morsels. That meme was passed on to others and it spread to their successors, modern chimps.

Although the LCA has been extinct for millions of years (if one came back to life he or she could not successfully cohabit with any chimp or human alive today) many (perhaps most) of our modern chimp and human memes are more or less identical to those the LCA modified during its time on Earth and passed on to us.

Point #2 The memes invented by we humans, particularly the memes of metaphoric language and the flood of religious and technological and scientific memes that came out of that meme will be passed on to our successsors. Thus, even if humans go extinct, if our cyborg successors have many of the same genes we modified and passed on, we will not "really" be extinct. Our MEMORY will still be in the cyborg memenome.

Ira Glickstein

Ira Glickstein said...

Howard, I agree "fictional story telling is the first exclusive human function of language, and it is still a dominant function." [Emphasis added - I agree most stories are fictional, in the broadest sense of the word.]

However, I disagree with you (and Joel) when you say you "see no obvious adaptive value in most stories." I believe most stories are adaptive in the broad context of that word, including the fictional stories!

What do we mean by "fictional" - we mean not literally true. The "holy scriptures" of all religions convey:

(a) Literally true facts (e.g., a man named "Jesus" lived and preached in the vicinity of Bethlehem and Jerusalem in the first century BC).

(b) Untrue assertions (e.g., Jesus was the Son of God).

(c) Fictional but containing deep and valuable truths (e.g., Jesus's teachings about how humans should live and interact with each other, a faith in a force for universal good, ...).

The same triage may be performed on Aesop's Fables, Darwin's Origin of the Species, Newton's Laws of physics, Einstein's Relativity Theory, the US Constitution, today's newspaper, and almost any fiction or non-fiction book on almost any subject.

Why do I think these (mostly) fictional works are adaptive? Well, just as "a little bit of honey makes the medicine go down" a bit of fiction makes deep truths more palatable and MEMORABLE.

George Carlin was a comedian but he made some social and political points we remember because of his fictional add-ons and presentation.

Einstein created some fictional cases (the man sealed in an elevator, a man travelling alongside a light beam, ...) that illuminated some deep truths of physics for himself and us.

Is physical exercise adaptive? I would say most of it is, to keep our muscles and bones healthy. Some teens take exercise to excess by playing basketball when they should be studying, but others do even worse by spending their time playing videogames (not Wii :^) instead of studying.

Similarly, I think making up fictional stories and enjoying the stories of others is adaptive in that it keeps our minds active and interested and helps us absorb information that will be useful in planning our future activities.

Ira Glickstein

Howard Pattee said...

Ira, I think we differ only on the word “most”! You are picking special cases that look adaptive. That is not “most” of recorded language. Statistically speaking, most of the books in the library rarely circulate. Most of the language on TV is a tale told by idiots, signifying nothing. Most of the text in the memory of my computer will eventually be deleted or lost. Most religions are now extinct, and I doubt that most of those that presently exist are adaptive.

I think we may disagree substantially on the adaptability of cyborgs, but that requires another post. I have to think about it.

Howard

joel said...

I've been thinking about the nature of memory and its relationship to tropism, trying to invent a mechanism that seems reasonable. Try this one on for size. The simplest single cell creatures are opportunistic in their feeding, however, some are capable of moving via cilia toward their food supply. That "attraction" may actually be a sensitivity to gradients in chemistry or electric charge. We know that gradients are essential to the functioning of a single cell and the growth of tissue. Theoretically, gradient can be determined by measuring an environmental property at opposite ends of a single cell. Another method, capable of detecting even smaller gradients, is to measure the property, remember it, and then to move in a random direction to make a second determination. Casual observations that I've made under the microscope suggest that the latter method may be actual. The motion of some cells seems like random walk plus a drift. This would be the most elemental use of memory. All that is necessary for memory is a measurement, storage and a time delay during which random motion to a different position takes place. The storage can be actual or osmotic ingestion of the surrounding material.

To visualize this a little better, imagine a sponge made of an insulating material immersed in a saline solution. After the sponge is saturated, we move it to a solution of a different salinity. The outermost layer of the sponge will quickly come to the salinity of the surroundings, but the interior will remain at the initial salinity for awhile. The initial salinity is "remembered." The salinity gradient in the sponge could run a motor to propel the sponge to another location.

While running this though experiment the usual question occurred to me. (This is the philosophical part.) Who cares? Suppose I've got this exactly right. What does this tell us about anything that's important to us? Suppose I could via speculation follow the evolutionary path right up to memory organization in human, does this tell me anything about the human condition?

With respect -Joel

Ira Glickstein said...

Memory in single-cell organisms is an interesting issue and I'm pleased Joel brought it up.

Joel is not talking about the DNA memory inherited by the single-cell organism from fit predecessors, but some kind of near-real time memory process within the single-cell organism itself. Howard, for example in this academic paper, makes a distinction between what he calls "rate-independent" and "rate-dependent" processes.

DNA memory, brain memory, and computer memory are examples of "rate-independent" processes. The information is recorded and stays in memory for an arbitrarily long period of time. It does not matter how long it takes to access the information.

According to the linked paper, "...physical theory is described by rate-dependent dynamical laws that have no memory, ..."

Joel suggests something that seems to be in the middle, and I would love to hear Howard's reaction.

Joel imagines a single-cell creature that requires some kind of nourishment that is distributed in the water environment as a gradient (that means there is less of it in some areas and more in others). The single-cell organism absorbs some of the water and it passes through to an inner partition (call that the "memory"). It then moves in a random direction and absorbs some more water into an outer partition and, using some physical process, compares the concentration of the nourishment between the "memory" sample and the new sample. If the new sample has more nourishment, it continues moving in that direction, else it moves randomly in another direction. The type of "memory" Joel imagines is by its very nature dynamic and certainly rate-DEPENDENT.

Well, Howard, is it or is it not memory?

Ira Glickstein

PS: I know that some single-cell organisms have cell walls that are chemically sensitive to the outside environment. (That chemical sensitivity is inherited in the DNA of the organism.) If the material outside the cell wall is toxic, the cell wall becomes less porous to keep it out. If the material outside the cell wall is nourishment, the cell wall forms an invagination that captures the material and allows it to be absorbed by the organism. I suppose, as Joel suggests, if there was more nourishment sensed at one end of the organism than at the other, some chemical reaction could cause the organism to move in the appropriate direction. However, that would be an example of rate-independent memory stored in the DNA.

Howard Pattee said...

There are many forms of dynamic or time-dependent memory like hysteresis and delay lines, but their capacity is strictly limited, as is their rewritability (recursiveness). A good random access memory must be relatively stable and have open-ended information capacity. Physically, that means it must have many relatively stable states with the same energy (energy degeneracy). Otherwise, over time physical laws would favor the lower energy states over the higher energy states. It also should be expandable and easy to construct from simple parts.

Linear sequences made up of a small set of distinguishable but similar elements is the simplest structure that satisfies these conditions. Genetic language and almost all of our natural and artificial languages are linear strings from a small alphabet. Strings are energy degenerate as long as the elements do not interact except in the one string dimension. Of course the degeneracy is broken when the string folds to make functional enzymes, which is why he whole genetic system works.

Two-dimensional arrays are useful in computer memories, but their construction and how you read and write is more complicated, and increasing their capacity is more difficult. The brain uses higher dimensional networks, but it is speculated that it may involve polymer sequences for long-term memory.

Many such copolymer sequences occur naturally, and for the origin of life problem assuming the simplest possible potential memory structures is the best hope.

joel said...

Howard Pattee said...

There are many forms of dynamic or time-dependent memory like hysteresis and delay lines, but their capacity is strictly limited, as is their rewritability (recursiveness). A good random access memory must be relatively stable and have open-ended information capacity. Physically, that means it must have many relatively stable states with the same energy (energy degeneracy). Otherwise, over time physical laws would favor the lower energy states over the higher energy states. It also should be expandable and easy to construct from simple parts.

Linear sequences made up of a small set of distinguishable but similar elements is the simplest structure that satisfies these conditions. Genetic language and almost all of our natural and artificial languages are linear strings from a small alphabet. Strings are energy degenerate as long as the elements do not interact except in the one string dimension. Of course the degeneracy is broken when the string folds to make functional enzymes, which is why he whole genetic system works.

Joel responds:

There's no arguing with the qualities that Howard requires for "a good random access memory." However, I do question that the evolutionary path to memory had such a memory as its goal. I think that what Howard calls "delay line memory" was the first possible step. A circulating ring memory would be the next plausible step, in other words, a delay line looped and connected to itself. I doubt that the human mind should be thought of as random access memory. Computer memory is addressable and the addresses must be stored in the program which is itself in memory and in turn must be addressable. I'm skeptical that Mother Nature uses this architecture. "Like a circle in a spiral, like a wheel within a wheel," a circulating delay line with interconnects between rings needs no addresses. Everything is always connected to everything else in a sense. Data is accessed by a search against template of all memory.

I'm bothered by the idea of dna as RAM. I think it gives us good reason to look at process memory (how to ride a bike) versus episode
memory (what I had for lunch). Dna contains no information about the past. Of course, it (like everything else) is a product of the past, but it is a set of instructions about the present or future. Like never forgetting how to ride a bike or the difficulty of breaking a habit, dna has a fixed way of behaving unless it undergoes physical damage or the trauma of miosis. Should we think of it as a program rather than writable and extensible memory?

With respect -Joel

Howard Pattee said...

Joel is bothered by the idea of DNA as RAM. I agree it is like a memory-stored program. I was thinking of development where apparently any structural gene can be accessed (turned on or off) whenever necessary. (Random access is a misnomer. Arbitrary access isn’t quite right either.)

Joel also says DNA has, “no information about the past,” although like everything else, “it is a product of the past.” He speaks of DNA as a set of instructions “about the present or future.”

Here is the way I look at it. What we call “the past” and “the future” are strange concepts because both exist (i.e., can be directly accessed) only in our present memories. All we believe about the past and the future are only inferences based on our present brain states and predictive models.

Of course as Joel says, all memories (records, histories, images, models, measurements, data, information, etc.) are “products of the past.” There is no “memory” of future events, by definition.

My point is that whether or not “products of the past” have “information about the past” depends entirely on how they are interpreted or “decoded.” It happens that the most accurate information about the past tree of life comes from the statistics of present protein and DNA sequences.

Ira Glickstein said...

I thought I was the only one alive who remembered delay lines but both Howard and Joel brought them up!

(For you youngsters, a delay line is a device used in the olden days for analog signal processing. A modulated signal is introduced at one end and it passes through the delay line in a period of time. There are taps on the delay line that provide samples of the signal separated by pre-set times. By comparing the outputs at the various taps, you can determine periodic characteristics of the modulated signal. I once did a conceptual design of a giant stack of delay lines but it was never built. Nowadays, this type of signal processing is done in digital computers.)

In thinking about Joel's and Howard's ideas about delay line-type memory, I naturally thought about human short- and long-term memory.

It appears that all our memories are first recorded in a part of the brain designed to hold memories for only short periods of time, after which they dissipate.

That is useful for remembering a phone number from the time you read it in the phonebook till the time you dial it. After that short period, it is of no use and should be discarded - and it is!

Of course, some stuff needs to be remembered for longer times - perhaps for your whole life. That stuff needs to be transferred from short-term emory to long-term memory where it stays for arbitrarily long periods of time - perhaps till we die.

By chance, I happened to see a this Scientific American story (July 2008) about the transfer of information from short- to long-term memory and how it may be disrupted in older people.

"In a study of rats, the scientists found that when the animals were at rest there were repeating patterns of neuronal (nerve cell) activity believed to be involved in moving information from short-term to long-term memory vaults in the brain. The process, however, was disrupted in the older rats.

Now I know why I had to dial my cell phone from the house phone to find it this morning! (It was in the golf cart in the garage :-(

Ira Glickstein

joel said...

Here's a quirky little idea that interests me (at least a little bit).

Howard said, "Of course as Joel says, all memories (records, histories, images, models, measurements, data, information, etc.) are “products of the past.” There is no “memory” of future events, by definition."

Of course I agree that there is no memory of future events and yet.... Virtually every science fiction story, including Ira's, takes place in the future. However, the narration is in the past tense not the future tense. That means the narrator exists sometime in the far future and is talking to us through some kind of time warp. That seems a bit bizarre.

I would speculate that in the history of human story telling recounting tales of past adventures of heros or talesof how things got to be the way they are in the present, was the only material they had to work with. A shaman or soothsayer would tell what was going to happen in the future tense. The future tense may even have been forbidden to the common folk. Evidense for this is that even today speaking of the future may be taboo in some cultures for fear of inciting the "evil eye." You would think that a special grammatical tense might have been evolved, but science fiction is too new for that, and besides we all accept the convention of the future-narrator speaking of his time's past. To emphasize that things might have been otherwise, I point out that French has a special tense, the "passé simple" which is used for story telling of the far past. Also, if I recall correctly, Japanese uses no verb tense at all, depending on context to convey the time period. And, speaking of odd verb forms; I think I shall pour myself a glass of wine. With respect -Joel

Ira Glickstein said...

Joel, I agree it is a bit bizarre for the narrator of a story that takes place in the future to speak in the past tense as if he is in the far future.

As you point out, that is exactly what I do in my novel, and it is the current convention -in English at least- for stories of that type.

During a period where they have heard gunshots and are in danger of being killed, my traditional Christian character says to my non-believer character "Don't worry about the future, God is already there!" (Chapter 6)

Since I, as the author of the novel, control all the action, I am -in essence- the "God" of that Creation. I am already "there"!. So, I guess I have the right to speak in the past tense!

Ira Glickstein

Howard Pattee said...

Linguists call this bizarre property of language that allows the author to appear to be anywhere at any time, and to describe events that may or may not occur anywhere and at any time “displacement”. It used to be argued that displacement is what distinguished human language from animal language, but this is highly controversial. Animal studies show all kinds of intermediate examples. We really don’t know what is going on in animal brains, or even our own brains.
http://en.wikipedia.org/wiki/Animal_language

Ira Glickstein said...

Thanks Howard for the link to Wikipedia - "Animal Language".

They say:

The following properties of human language have been argued to separate it from animal communication:

Arbitrariness: There is no rational relationship between a sound or sign and its meaning. (There is nothing "housy" about the word "house".)

Cultural transmission: Language is passed from one language user to the next, consciously or unconsciously.

Discreteness: Language is composed of discrete units that are used in combination to create meaning.

Displacement: Languages can be used to communicate ideas about things that are not in the immediate vicinity either spatially or temporally.

Duality: Language works on two levels at once, a surface level and a semantic (meaningful) level.

Metalinguistics: Ability to discuss language itself.

Productivity: A finite number of units can be used to create a very large number of utterances.


I agree with Howard that animal communication has much in common with human communication, and that there are logical arguments to dispute some of the above "human-only" properties.

However, even the communications of our nearest relatives, the Chimps, exhibits only the faintest shadows of shadows of human-level metaphoric language.

As for "displacement" - ability to "communicate ideas about things that are not in the immediate vicinity either spatially or temporally" which Joel thought was "bizarre" when a storyteller uses past tense to describe an event in the future by placing him- or herself in the further future, I do not believe there is any such thing in animal language.

The "bee dance" that effectively communicates the distance and direction and quantity of a food source is what I would call a "sign". If the bee said "go northeast at 34 degrees, 1100 yards, to a half-acre field of clover near a big oak tree" THAT would be symbolic language. But that is not what the bee does at all. The bee turns and steps in particular directions with particular distances and intensities to convey the information as a set of analog signs that are directly related to the message and not at all arbitrary.

Even after reading the item linked to by Howard, I retain my belief that all animal communications we are familiar with, including bees and chimps (and probably dolphins, though I am not sure), are signs that can only denote warnings, threats, and requests.

The bee dance is of the nature of a request, "fly in this direction a certain distance and harvest a big store of pollen." All the other examples are warnings of predators, threats to conspecifics or predators to back off, or requests to conspecifics to help search for or obtain food or other resources in cooperative hunting.

When an animal, such as a dog or horse, responds to arbitrary human words it has been trained to obey, that is in the nature of a request with an accompanying threat that, "if you don't do it I'll whip you or not give you a reward." Similarly, chimps that use keyboards and such to respond to human questions are using their request and threat communications capabilities superimposed on human-created language that they only understand peripherally.

I'll go further and claim that, prior to the invention of metaphoric language by humans some 6,000 to 10,000 years ago, even humans were limited pretty much to warnings, threats, and requests.

I'll grant that in the ten-thousand years prior to the invention of metaphoric language there were some very complex and subtle combinations of warnings, threats, and requests used and understood by humans, but it was not at the level of metaphoric language.

Ira Glickstein

Howard Pattee said...

This discussion brings me back to Biosemiotics.

There are seven properties that conservatives claim elevate humans above lower animals. They are viewed liberally by biosemioticians as basic conditions necessary for all life. That is, viewed not as conservative anthropocentric properties, but as liberal generalized functions, they are the necessary conditions for open-ended evolution.

Here are the 7 narrow human conditions followed by their molecular genetic equivalent functions.

Arbitrariness: There is no rational relationship between a sound or sign and its meaning. (There is nothing "housy" about the word "house".)

Cultural transmission: Language is passed from one language user to the next, consciously or unconsciously.

Discreteness: Language is composed of discrete units that are used in combination to create meaning.

Displacement: Languages can be used to communicate ideas about things that are not in the immediate vicinity either spatially or temporally.

Duality: Language works on two levels at once, a surface level and a semantic (meaningful) level.

Metalinguistics: Ability to discuss language itself.

Productivity: A finite number of units can be used to create a very large number of utterances.

Here are the genetic language properties:

Arbitrariness: There is no rational connection between a base triplet of DNA and the amino acid that it stands for. The entire genetic code is chemically arbitrary. Monod calls this the “principle of chemical gratuity” that is characteristic of all enzymes and proteins. Innumerable gene sequences could provide the same functions, just as innumerable languages and sentences can mean the same thing.

Cultural transmission: Genetic language is also passed from one generation to the next. Cultural transmission is just less reliable.

Discreteness: The genetic alphabet is also discrete consisting of 4 bases and 20 amino acids. Most of the 6000 languages have about the same number of letters in their alphabets.

Displacement: Symbol reference and symbol function must not be confused with symbol writing and reading. All human writing and reading is done only in the present. The thoughts these symbols induce in the brain also exist only in the present. The same is true for protein synthesis and enzyme catalysis. Most proteins are produced for future functions. That is what development is all about. In fact the cortex of the brain is constructed only for future use (after learning).

Duality: Genes also work at many levels at once. Self-replication requires copying DNA molecules without interpreting them, then interpreting the sequences by a code, and finally integrating their functions or meanings.

Metalanguage: Genes have a complete metalanguage that is essential for development. That is, genes can turn themselves off and on. In fact genetic metalanguage is much more complex than human language.

Productivity: Genetic language can produce an endless number of sequences, just like humans. In fact it is only because of genetic productivity that humans and their languages exist.

Ira Glickstein said...

Excellent Howard! I love your last Comment!

I agree that the genetic language is a true language, in the same class as human languages.

Of the seven properties, you have provided convincing arguments for six - all except "Metalinguistics: Ability to discuss language itself."

Is the complex system in the genetic language for turning genes on and off really analogous to our use of human language, our "Ability to discuss language itself?"

Our human capability for self-reference is what many scholars think is behind what we experience as consciousness.

Is gene control and regulation somehow equivalent to Metalinguistics of human language or self-reference?

I would love to hear your arguments (if you have any) in favor of that equivalence.

If the genetic language is capable of self-reference, it, as a system consisting of all living and interacting biomass on Earth, may be Conscious at a level above that of any single animal!

Another question: Is the genetic language at all capable of something like metaphor?

I would love to hear your arguments (if you have any) in favor of that equivalence.

Ira Glickstein

Howard Pattee said...

I stated only the basic functional similarities between genetic and human language, but of course there are also big differences, as you would expect after 4 billion years of evolution! How neural networks in the brain produce symbol sequences of language is a mystery, but it is certainly unlike the production of genetic sequences. It would be an anthropomorphic stretch to imagine figures of speech, metaphors, irony or humor in genes.

One big difference is how “displacement” works. Human languages have complex syntax and many parts of speech, nouns, verbs with tenses, and prepositions to explicitly express time and space relations. Gene syntax is extremely simple. The basic genetic controlled constructions have only one person, one tense, one mood, and one part of speech. All constructive gene expressions roughly correspond to the 1st person singular, present imperative, and use only nouns. Genes cannot make even simple declarative statements, ask questions, or make propositional statements than can be true or false.

Genetic sequences for structural proteins are made up of instructions of the form, “Add part X here and now!” The X is like a proper name directly referring by way of the code to each of the 20 amino acids, and in copying to each of the 4 bases. None of the cell’s actions are expressed explicitly by verbs (“add” is implicit). All actions are the result of the constraints of these noun sequences on the physical dynamics. Consequently, it is the physical behavior of these actual constructions, the proteins not the DNA symbols, that influence their spatial and timing displacement.

However, the genetic metalanguage is essential, especially for development. Ira is correct, the metalanguage embedded in natural language is different. It says something about the text, but does not normally determine what sentences and paragraphs are read or omitted in a text.

By contrast, replication and development in organisms depends entirely on gene regulation ― the coordinated expression or omission of structural genetic sequences as determined by many control sequences that are themselves influenced by a complex network of interactions with the internal and external environment.

The much larger numbers of non-structural and non-coding genetic sequences form a network of control functions that are still being discovered. Some of them function syntactically like punctuation, but most controls involve many genes and have no apparent similarity to natural language syntax. For example, some DNA sequences are read in both directions and each product has a different function.

A stylistic difference between genetic language and cultural texts is that humans continually create new stories, and tend not to repeat existing texts. In fact, copying is not considered creative and plagiarism is unacceptable. There are also thousands of natural languages.

By contrast, genetic texts are essential evolutionary histories that are highly conserved if they describe a useful protein or control function. The major differences in species are the result of the metalanguage control of conserved gene expressions. Plagiarism is the norm, and there is still only one basic genetic language.

Ira Glickstein said...

Howard, you say "All constructive gene expressions roughly correspond to the 1st person singular, present imperative, and use only nouns. Genes cannot make even simple declarative statements, ask questions, or make propositional statements than can be true or false."

I agree gene language per se, cannot make statements or ask questions, etc.

However, neither can human language, per se! It is each individual human who, using human language, makes statements and asks questions, and these questions and statements only make sense to those humans who understand that language.

Let me suggest that the Earth's biomass uses the genetic language to make statements and ask questions. Some of the statements are of the form "this well-established and time-proven formula for making blood or bone or whatever should be highly conserved and protected from mutations, therefore the phrases of the formula are repeated in the DNA and are also protected by corrective mechanisms in the genome." Indeed, some of these highly conserved sequences are so well protected as established "truths" that identical genes are found in peas and humans!

Throughout human civilization, we have passed down well-established truths to the next generation in the form of statements in language: "honor your mother and father, work hard, don't steal, ..." and so on. Despite the rapid development of technology, human nature remains pretty well conserved over thousands of years.

As for asking questions, it seems to me that the crossover in sexual reproduction and the mutation rate in all reproduction is similar to asking questions of the form "what would happen if fish fins were modified a bit so they could be used to propel the fish on the bottom of a muddy river and eventually on land, and if gills were modified a bit to extract oxygen from air as well as water?"

Much human invention involves crossover and even mistakes. The Wright Brothers observed the function of wings in birds and conbined that with technology from bicycles and internal combustion engines and created the first airplane. Pennicylin was accidentally discovered, etc.

I also take issue with your claim that the genetic language includes only nouns. It is of course true that the genetic language, per se, is simply a parts list and description for construction of proteins. However, if we stand back a bit and consider the reproductive system as a whole, it is clear there are at least implied verbs that cause actions necessary to constuct hearts and eyes and other organs out of these proteins. Verbs in human language are merely names of actions. Indeed, all parts of speech are merely words that name a thing or an action and so on. If I say "Ira[noun] is a bicyclist[noun]" that is the same as "Ira[noun] rides[verb] a bicycle[noun]".

The complex gene regulation and control mechanisms that are activated in the genetic reproductive system are the result of interpreting the gene sequences in DNA.

Another point you make is that there is only one genetic language on Earth. That is true, but there are some biological organisms that make use of a few amino acids not in the human genome and vice-versa. These are like "dialects" of a human language. Of course, the genetic language has been in development for some three or four billion years, Homo sapiens language for only 100,000, and metaphoric language for less than 10,000. International commerce has pretty much settled on English as the standard and I expect within the next century virtually all people on Earth will speak and understand a common language.

Ira Glickstein

Howard Pattee said...

Of course Ira is correct that any symbol vehicle per se is utterly meaningless until there is an interpreter. Semioticians agree that the concepts of sign and symbol imply an irreducible triadic relation: symbol-interpreter-referent. The interpreter is by far the most complicated, largely arbitrary, and usually ill-defined part of the relation. There is usually ambiguity in where the signs and symbols stop and the interpretation begins.

For example, when you hear the word “mosquito” your auditory nerve is not interpreting but just carrying pulse-coded symbols to your brain, the same as when you see a mosquito your optic nerves are just carrying symbols. When and how the brain becomes an interpreter instead of just a symbol processor is not understood. At least when you swat the mosquito we can be sure your brain has made an interpretation. Note that you may be busy at some other task and virtually unconscious of this complicated process.

In order to make the meaning of symbol more precise, I have limited it, by definition, to a structure that is not determined by energy-based physical laws. That just means you can string nucleotides, amino acids, words, or bits together in any order without worrying about their energy requirements. (This is called energy degeneracy.) However, whenever physical laws take over, as in the folding of proteins and subsequent enzyme catalysis, or swatting a mosquito, we are no longer in the symbolic domain, but the energy-based domain.

That is why by my definition of symbol, a gene does no more than describe the structure of proteins. There is no genetic symbol or “verb” that describes the folding or action of the protein. To “imply” verbal action is a human doing the interpretation. Natural language does have symbols for actions, and they can be manipulated symbolically without any energetic action taking place. A symbol for action is not an action.

You suggest that the Earth's biomass uses the genetic language to make statements and ask questions. According to modern evolution theory, the only process by which the Earth influences genes is by natural selection. I think describing natural selection as the Earth making statements is a fine anthropomorphic metaphor, but I don’t see how it could be a testable scientific hypothesis.

Howard Pattee said...

In answering Ira’s disagreements I discovered a logical inconsistency between two of my posts.

I said that genes as symbols can only describe the structure of proteins, and that genes have no symbols (no verbs) to describe their later actions. I think this is correct by my definition of symbol.

But then it follows that the genetic language cannot describe displacements in space and time. I don’t know of any genetic symbols corresponding to verbs and prepositions that in human language allow it to describe displacements. So I conclude that displacement is still a property that distinguishes genetic language from human language. However, I think Ira’s disbelief (7-31-08) that, “there is any such thing [as displacement] in animal language” is empirically questionable. I’ll try to find some counterexamples.

Howard