Information Theory Continued
We have settled upon a working definition of “information.” It is: Semantical content intentionally formed and conveyed according to syntactical rules.
We see three primary components of this definition: Syntax (form/structure), semantics (meaning), and intention. Each is a necessary condition for a particular “state of affairs” to be information.
An evolutionist might now object, “Wait just a minute! You are loading the game in your favor by talking about ‘intention.’ Intention implies intelligence, so you are saying that there can be no information without intelligence. But we know that there has been information without intelligence from the inception of life. So your definition cannot be correct.”
And we reply: “You are right that information implies intelligence. Virtually all scientists recognize this fact. Indeed, the SETI project itself recognizes and depends upon this fact. So, we are not incorrect to recognize this fact also. The problem you face is to explain how inanimate, non-intelligent sources produced intelligence and information.”
So, we stand by our definition, including the necessary condition of intelligence. And the best response an evolutionist can have is to say that intelligence and information “emerged” in lockstep fashion, with a smidgen of information “emerging” alongside just enough intelligence to be able to “make something of” that smidgen of information. Then the information got a smidgen more “complex,” as did the necessary intelligence to “make something of” that additional information. So, the evolutionist must imagine a bootstrapping process in which complexity of information “emerged” in lockstep with the requisite intelligence able to “read” and “understand” that information.
Let us now consider all three necessary components of information and show why the best evolutionist model of the emergence of information is impossible in principle.
“Syntax” refers to the form or structure of an entity. Thinking of sentences as entities, we can see how the form of sentences can affect their informational content. Consider the following two sentences that are composed of the same words:
“The door is shut”
“Is the door shut”
The mere ordering of the words, even in the absence of punctuation, affects the meaning. The first sentence is a statement, while the second is a question. Your mind even supplies the missing punctuation, because you recognize the syntax and what that syntax typically signifies.
Our SSL key example from our last lecture shows that there can be syntax even in randomness. The syntax is supplied by the rules of the SSL algorithm; it is not “in the key itself,” so to speak. So, in the SSL protocol, both the random string of characters and the rules of the algorithm work together to produce the syntax of the string.
But, just a minute. That is exactly what’s going on with all “patterns,” even the ones that don’t appear random to us. For example, to a non-Arabic speaker/reader, a sentence of Arabic can appear like random marks. Even the “sentence order” is not what Westerners expect, as Arabic is written right-to-left instead of left-to-right. And there are many other examples. Recognizing the patterns in a language is a function of “knowing the language,” but that is in turn a function of “knowing the rules,” which just is “knowing the algorithm” (slipping into SSL-speak).
The fact is that syntax is not trivially “a pattern.” Recognizing “states of affairs” as patterned just is the imposition of rules. So the rules have logical priority; patterns imply rules. And the rules vary based upon the “game” you are playing when you contemplate the “states of affairs.”
This is a fundamental feature of what is wrong with the SETI project: It presupposes that we are going to be able to find some “rules” according to all the “games” we know how to play among the seemingly random “states of affairs” that come streaming into our collectors. But we actually have no reason to think (and in fact should presume) that any alien intelligence will be communicating according to rules of games that we cannot even at present imagine. Worse, to recognize as information the stream of randomness, we must already know the syntactical rules. We don’t get the rules from the random stream. Instead, the random stream can only “be something to us” if we already know the rules.
The problem of a “state of affairs” conveying something is hard enough when humans are trying to learn a new human language. And that’s with the help of a teacher acting as a “translator” (which just is teaching the rules of the new “game”). See this Wikipedia article that is a reasonably good synopsis of Quine’s “Radical Translation” article.
But the problem is infinitely magnified when trying to understand a totally alien, non-human language, and that without a translator on hand to teach us the rules of the new game!
The point is that syntax is always imposed on “states of affairs” according to pre-established rules. The “states of affairs” themselves don’t have any syntax built into them. We see patterns because we impose them, and we do so according to pre-established rules. There is nothing inherently “ordered” about this very sentence I am now writing! The “order,” the syntax, you see in it emerges because we are both English-speakers, so I am intentionally writing according to the rules of English, and you are reading according to the rules of English. But if you are not an English-speaker, then this sentence I am presently writing will be meaningless to you, and that primarily because you don’t know the syntactical rules.
The rules make any particular syntax what it is. The rules do not emerge from pre-existing syntax; they cannot, as there is no empirical mechanism for that to occur. Instead, we impose rules (of games we know) onto the “raw empirical data” and thereby see syntax that is not “in the things in themselves.”
And now you should be seeing the brilliance of Kant once again. The “categories,” as he calls them, just are the rules of human thought (and perhaps some animals as well). These provide the basic structure of thinking about anything, including additional, derivative rules of new games. You cannot “build up” rules of new games from nothing whatsoever. Instead, like the axioms of logic that can then be built up into the vast and robust logical schema we now recognize, all of linguistics is rule-based where the rules of all natural languages are built up from the axioms of the “categories.”
The rules logically precede syntax. Syntax is what it is to us in any given context only when we know the rules of that context. Syntax is a necessary condition for information, and pre-established rules are a necessary condition for syntax. So, the existence of information necessarily presupposes the existence of rules.
At this juncture, an evolutionist might still respond, “Okay, okay. But the ‘rules’ you are talking about are themselves just further physical relations. We recognize a salt crystal as a ‘cube,’ but that arrangement, that ‘syntax,’ is what it is because of the laws of chemistry. Those are the ‘rules of the syntax of salt crystals,’ if you will. And the same principle applies to all physical patterns. As we look deeper and deeper, we discover more and more of the ‘rules of syntax’ that produce the patterns in nature that we recognize. It’s all natural law working on matter, and strictly speaking we must be careful to not slip into anthropomorphizing either the ‘rules’ or the ‘information.’ Both are really just material states of affairs.”
We will respond in two ways.
First, we do not agree that all patterns count as “syntax” in the information-relevant sense. A salt crystal’s cube shape is no more “information” than is a natural pattern of growth on an island that we recognize as “HEL….” If the arrangement is believed to be an entirely natural occurrence without any intentional meaning, then we don’t take that arrangement to be conveying information. Again, SETI itself operates on this very principle!
Second, the very fact that we can distinguish between a naturally-occurring pattern and one designed to convey meaning indicates that we don’t necessarily anthropomorphize every pattern we happen to detect. We can say, “This one is a cry for help, while that other one conveys nothing because it is not information.”
Syntax is a necessary but not sufficient condition for information. Semantical content is also necessary.
“Semantics” is the study of meaning. Certainly, as we saw just above, syntax is an aspect of meaning; syntax is a necessary but not sufficient condition for information. However, there is much more to semantics than just syntax!
If ever there was a philosophical subject that we will barely touch upon, this is it: philosophy of language. This field of inquiry is so vast that all I can do is point you in some directions for further research on your own, while telling you what I, most other philosophers, and many scientists have come to realize about meaning: Just as syntax is what it is because of pre-established rules, semantical content is what it is because of abstract objects called propositions.
You’ll remember Rudolph Carnap, who we talked about as one of the “big guns” in the Vienna Circle, one of the fathers of Logical Positivism. Carnap was a hard-core materialist-empiricist, and he believed that empirical truth-conditions are the anchor of semantics. You’ll remember that the core doctrine of Logical Positivism was: “A sentence lacking empirical truth conditions is worse than merely false; it is actually meaningless.” And you’ll remember how Logical Positivism foundered on the discovery that its core doctrine is therefore meaningless, as it itself lacks empirical truth conditions.
I remind you of Carnap, because I want you to put the following quotation from Carnap in its proper context and feel the force of what he, as an empiricist about semantics, is saying. In a 1950 article, entitled “Empiricism, Semantics, and Ontology,” Carnap writes:
Empiricists are in general rather suspicious with respect to any kind of abstract entities like properties, classes, relations, numbers, propositions, etc. They usually feel much more in sympathy with nominalists than with realists (in the medieval sense). As far as possible they try to avoid any reference to abstract entities and to restrict themselves to what is sometimes called a nominalistic language, i.e., one not containing such references. However, within certain scientific contexts it seems hardly possible to avoid them.
Now, “nominalism” is a metaphysical position that certainly takes Occam’s Razor to heart, which you’ll remember can be summarized as: “Don’t multiply entities beyond necessity.” So, nominalists want the most minimal ontology (set of what really exists) possible. For a reasonably good summary of the position, you can look here. However, we can nutshell the view to say that it denies the existence of “universals” and abstract objects (entities that do not exist in space and time). The idea is that reality consists of only particular things in space and time. So our use of universal terms (properties like “just,” “right,” “tall,” “green,” and so on), and references to abstract objects like triangles and numbers are all “built up” or fictional terms that do not actually refer to real entities. Such terms are just ideas that we have, and we use such terms like they refer to things in the world. But they really just refer to ideas we have. Only are particular terms refer to anything in the real world, because the real world is composed of all and only particulars. There is “this particular cat,” but there is nothing in the world that is the “property of catness.” There is “this particular shade of green,” but there is nothing in the world that is the “property of greenness.” And so on.
Empiricists, like Carnap, are very fond of nominalism, because (like Hume and Locke) empiricists today really think that the world is filled with all and only particulars, and that we therefore only have particular ideas that arise from particular experiences. You never do experience “catness” or an ideal triangle. You only experience this or that particular cat. And your experience of “triangles” is always of this or that particular triangle. We never experience universals nor abstract objects. So our use of universal and abstract terms cannot be referring to real entities in the world.
This, however, produces what has come to be called “the problem of universals,” namely that empiricists must give an adequate account of how such ideas ever emerge in us. After all, we do think in terms of “catness” and “greenness” and so on. We do think that the Pythagorean Theorem applies perfectly to ideal right triangles and only more or less crudely to our crude particular representations of a right triangle. How do we get these universal and abstract ideas, when all of our experiences are of particulars that are in space and time?
And for our purposes in this section on semantics, the more pressing way to cast the problem is: How do universals and abstract terms come to have actual meaning, when our empirical ideas are always of particulars in space and time?
This latter sense of the problem is the one Carnap is referring to, and, again, remember that Carnap was both an empiricist and Logical Positivist. In this context, let’s continue with what Carnap says:
Recently the problem of abstract entities has arisen again in connection with semantics, the theory of meaning and truth. Some semanticists say that certain expressions designate certain entities, and among these designated entities they include not only concrete material things but also abstract entities e.g., properties as designated by predicates and propositions as designated by sentences. Others object strongly to this procedure as violating the basic principles of empiricism and leading back to a metaphysical ontology of the Platonic kind. It is the purpose of this article to clarify this controversial issue.
Notice that Carnap is going straight at the semantical aspect of the problem of universals and abstract objects. We use these terms. We think they are meaningful. But empiricists deny that these terms refer to anything in reality. So, wouldn’t it be better to just be systematic (and more accurate) in our language to stop using such “fictional” terms?
Carnap actually argues otherwise:
… it will be shown that using such a language does not imply embracing a Platonic [belief in the real existence of universals and abstract objects] ontology but is perfectly compatible with empiricism and strictly scientific thinking. Then the special question of the role of abstract entities in semantics will be discussed. It is hoped that the clarification of the issue will be useful to those who would like to accept abstract entities in their work in mathematics, physics, semantics, or any other field; it may help them to overcome nominalistic scruples.
Please note how Carnap contrasts Platonic with nominalistic metaphysics. The former believes in the real existence of universals and abstract objects, while the latter denies real existence to these fictional entities. And Carnap wants to provide a means by which these terms can be considered meaningful and useful, while with nominalism denying that they refer to anything in the real world. He wants to support nominalism, but he also believes that such terms are meaningful and useful. So, the question is: How can such terms have meaning, when they don’t pick out anything in the real world?
After a lengthy discussion about semantics as it was understood in Carnap’s day, Carnap concludes that not enough is known about semantics to see how a purely nominalistic ontology could ground meaningful universal and abstract terms. You might imagine that Carnap would be quick, then, to advocate discarding their use. But he does not:
Among those philosophers who have carried out semantical analyses and thought about suitable tools for this work, beginning with Plato and Aristotle and, in a more technical way on the basis of modern logic, with C. S. Peirce and Frege, a great majority accepted abstract entities. This does, of course, not prove the case.
So, Carnap is saying that “a great majority” of philosophers until his time opposed nominalism and were “realists” about universals and abstract objects; Carnap states that nominalists are in the minority. Of course, the truth or falsity of a metaphysical position is not decided by majority vote. But Carnap’s point is that the “great majority” of people thinking seriously about the problem of universals have concluded that the problem can only be solved by admitting the existence of universals and abstract objects.
Occam’s Razor can hardly apply to such entities if they are indeed necessary for our everyday sentences to have meaning! And Carnap is quite pragmatic about the tension he feels as a nominalist empiricist, yet seeming forced to accept that empiricism cannot offer an adequate response to the problem. He is quite intellectually honest as he summarizes the state of affairs. Referring to nominalists as “nominalist critics” (critics of ontological realism in semantics), Carnap says that they have a hard row to hoe ahead of them:
… semantics in the technical sense is still in the initial phases of its development, and we must be prepared for possible fundamental changes in methods. Let us therefore admit that the nominalistic critics may possibly be right. But if so, they will have to offer better arguments than they have so far. Appeal to ontological insight will not carry much weight. The critics will have to show that it is possible to construct a semantical method which avoids all references to abstract entities and achieves by simpler means essentially the same results as the other methods.
So, nominalists might be right, but their response to the problem of universals isn’t adequate (up to Carnap’s time). It appears to Carnap that the more complex ontology including universals and abstract objects just might be necessary after all. But the jury remains out!
Meanwhile, what to do about those pesky universal and abstract terms? Systematically try to excise them from our language, so that our language is “empirically pure”? No, Carnap advocates against that “nominalistic purity.” He says that while we continue to study semantics, let us use the terms as much as they have value within any particular discourse:
The acceptance or rejection of abstract linguistic forms, just as the acceptance or rejection of any other linguistic forms in any branch of science, will finally be decided by their efficiency as instruments, the ratio of the results achieved to the amount and complexity of the efforts required. To decree dogmatic prohibitions of certain linguistic forms instead of testing them by their success or failure in practical use, is worse than futile; it is positively harmful because it may obstruct scientific progress.
So, Carnap says that if such terminology is useful and seems meaningful, let’s not get bogged down in semantics. Let’s just use the terms insofar as all people using them understand them, and we’ll worry about how they can possibly have meaning later!
Now, 60+ years of semantical research after Carnap’s article, where do we stand? Is it still true that “the great majority” of philosophers of language deny nominalism and assert that universals and abstract objects really do exist?
The answer is “Yes,” and all the more so. In fact, the most brilliant luminaries in philosophy of language have argued compellingly that meaning inheres in propositions and that translation is in fact impossible without appeal to these really-existing propositions as abstract objects. We will go into this aspect of metaphysics in more depth next week. For now, however, we can summarize that the majority of, including the most highly regarded, philosophers of language have utterly rejected nominalism and the empirical account of meaning. They argue that it is not the case that universal and abstract terms are either fictional or “built up” from repeated exposure to particulars.
It is not the case that particular things came together with other particular things and suddenly a meaning-laden pattern emerged. Such a model could only provide an account of particular terms, such as “Tom,” “this cat,” or “this boat.” It cannot account for everyday phrases like, “Tom took his boat out on the lake.” What is “boat” in general? What is “lake” in general? For that matter, what does “his” mean?
If I’m not looking directly at Tom’s particular boat (that you are literally pointing out to me) floating on a particular lake (that you are literally pointing out to me), I have no idea what you mean! And I cannot see or in any way experience the particular “possession” that “his” means! The particular terms have meaning, but when you talk about “boat” or “lake” or even “his” in non-ostensive fashion, the general usage of the terms ceases to have particular reference on strictly empirical grounds. Again, we will get into these arguments more next week.
For our purposes now, as we talk about information, we can summarize that the most convincing arguments state that the semantical component of information relies upon the existence of abstract objects, and these are incompatible with a strictly empiricist metaphysics.
When Carnap said that semantics was still in its infancy, we can now say that it is not. Philosophers of language have made huge strides, and we now realize that something like the Kantian model must be correct. There are pre-established “meanings,” just as there are pre-established “rules of syntax.” What we take to be “information” is what it is for us because we are already “wired” to recognize both syntax and semantical content in certain arrangements of “states of affairs” and not others. Our ability to distinguish between information and non-information depends upon pre-established, non-empirically-based rules and meanings.
So, even though Kant did not have an explicit philosophy of language, his epistemological and metaphysical principles have stood the test of time and scrutiny even in the field of philosophy of language. Things are what they are for us, and information is what it is for us, because we “package” these things according to pre-established rules and meanings that define the very parameters of what reality is for us. We are in a “box” of space and time, and that “box” is further defined by syntactical rules and propositional content. And both syntax and semantics transcend particular languages, modes of conveyance, and particular messages, because these pre-established rules are what make all of the particular experiences and information-exchanges possible in the first place.
But syntax and semantics are both necessary but not sufficient conditions for information. The final necessary piece is intention.
As we saw in our island example from our last lecture, information inheres in intentionality. A particular, even familiar pattern of things does not indicate that that pattern is actually information-laden. As we concluded last time, a pattern does not imply information, and the lack of a pattern does not imply a lack of information. The key feature of information that differentiates it from non-information is the intention to convey semantical content.
Now, of course evolutionists might be engaging in sloppy-talk as they refer to a DNA sequence as information-laden, when what they really mean is that DNA is nothing but a particular syntax with no semantics nor intentionality. After all, it’s far easier to consider DNA like a really complicated salt crystal, claiming that there really is no “information” in a salt crystal nor in a DNA strand. That way, evolutionists can treat the complex protein interactions based upon DNA to be akin to salt crystal formation: The correct conditions just cause ions to attract in purely deterministic fashion and form a crystal alignment that happens to be cube-shaped. So, perhaps DNA is much like that: The correct conditions just cause the bases to align in purely deterministic fashion to form a DNA strand that just happens to be “read” (talking very loosely now!) by enzymes to produce a vast spectrum of needed proteins. So, a salt crystal is not strictly “information,” and neither is DNA. There! Problem solved: There is neither semantical content nor intentionality implied by DNA; there is only syntax.
But wait. Problem not solved. DNA exhibits none of the deterministic, purely chemical-reaction features of salt forming into crystals. And evolutionary theory actually depends upon the fact that the protein relations in DNA are not deterministic; they can “mutate” to become quite different than they were before!
Each DNA codon can represent one of 64 different “states” (like a “letter” of an “alphabet”). Each of these “states” is a “rung” on the spiral “ladder” that composes a strand of DNA. A strand makes up a longer or shorter “sentence” composed of “words” that are differentiated from one another by a “letter” called a “stop codon.” So, literally, DNA exhibits every feature of natural-language sentences. Furthermore, neither the combination of the bases that make up a particular codon nor the ordering of the codons into “words” and “sentences” is deterministically arranged by the laws of chemistry.
Now, immediately, at least some evolutionists are going to say, “Wait! Wait! Wait! Surely you are not saying that DNA violates the laws of chemistry, that it does not follow the same principles as, say, the combination of elements into molecules! It certainly does. So DNA’s formation certainly is deterministic according to the laws of chemistry.”
And I will respond: “Now you are playing games with the term ‘deterministic.'”
It turns out that just having hydrogen and oxygen atoms together does not just “result” in water, as the atoms of the two elements “bump together.” It takes outside and additional energy (such as heat or electricity) to cause the electron orbits to conjoin into molecular water. So, even the syntax of a water molecule is not “as immediately deterministic” as, say, the syntax of a salt crystal.
Moreover, salt crystals always form the same way. The laws of chemistry determine the shape of a salt crystal, and there are not multiple possible “states” a salt crystal can be in. So, don’t be confused or intimidated by evolutionists pointing in triumph to so-called “self-organizing” entities, such as salt or other crystal structures. These structures are single-state and determined by the laws of chemistry. So, calling them “organized” is trivially-true and fails to explicate anything about genuine information. DNA is nothing like either salt crystals or molecular bonding in inorganic chemistry.
The empiricist problem of explaining the information content of DNA is much, much harder than any comparison with “self-organizing” structures make it seem! The “outside force” that puts the building blocks of DNA together is the action of enzymes, which are very complex protein molecules. And there is no “natural attraction” between particular bases that make up a DNA codon, as there is between even hydrogen and oxygen. Spark a couple of hydrogen atoms with an oxygen atom, and they immediately go “Okie-dokie” and jump together into a water molecule. But DNA works nothing like that! DNA must be assembled. The bases don’t just “jump together” in anything like “automatic” fashion the way inorganic molecules do. The enzymes can assemble the bases into codons any way they “see fit,” and the combinations work; there is no “natural affinity” between these complex proteins that make, say, the A and G bases jump together more often than not. The ordering does not “just happen” according to inherent chemical affinities. So, the assembling process of DNA is not “deterministic” in anything like the same sense as salt crystals or even molecular arrangements.
And that very fact is acknowledged by scientists, as they need the very non-deterministic aspect of the “assembly” process to “go awry” at times (like no salt-crystal ever does), so that the process does not produce perfect copies of DNA each time a cell divides. Evolutionists need there to be “mutations,” which just are non-deterministic “mistakes” in the “assembly” process of a DNA strand. It is the very fact that DNA is non-repeating and non-deterministic that makes it “interesting” and even possibly a vehicle for carrying information.
Now, what does this have to do with intention?
The ordering of DNA codons is neither random nor deterministic. It has both syntax and semantical content. It actually does contain coded instructions, as it contains the codes for building physical creatures. But that implies the intention to produce a physical creature.
Immediately evolutionists are howling: “Oh, right! Give me a break! Look, it’s a purely physical process with no evidence of intentionality anywhere to be seen. You creationists see ‘God’ everywhere, but the actual evidence is strictly of a purely physical process that ‘just works’ with no aspect of it ‘knowing’ anything at all.”
And that part about “knowing” is true, as far as it goes. Salt crystals don’t “know” how to form up. Enzymes don’t “know” what they are doing when they code up DNA strands and other enzymes use the code to produce the myriad of proteins needed to life to continue. All of the “parts” are quite apparently mindless. But that is missing the underlying point. To see the point, let’s switch to an example of a computer.
A computer is a vastly complex, physical machine. Neither it nor any of its parts are “aware” or are “getting” the information that the computer manipulates. The computer blindly manipulates information without understanding any aspect of what it is doing. The entire process is mindless and without any hint of intentionality.
Well, except for the intentionality that obviously inheres in the construction of the computer in the first place, and the information itself that the computer manipulates. When we see a complex machine that manipulates information, and we see it doing this according to a program, we immediately recognize the intentionality behind both the program and the computer that employs it to manipulate information.
Imagine that we knew nothing of DNA, and yet SETI was listening one day and started receiving structured but non-deterministic signals. Over time these signals were recognized to be a series of 64 possible “codons,” and we could even come to detect a “stop codon” in the code. We don’t yet “understand the message,” but we realize that this signal contains a definite syntax in the informationally-relevant sense. Do we discard this signal and just keep looking elsewhere for intelligence?
Certainly not! It is certain that SETI would immediately publish these results are “indications of extraterrestrial life” and then focus almost all energies on deciphering and understanding this signal!
With further research, SETI enlists the help of biologists who realize that if they start putting particular chemicals together in the sequences specified by the signal, with a particular chemical corresponding to a particular part of the code, a spiral helix molecule of astounding beauty takes shape. Scientists would then trumpet to the four corners of the Earth: “Amazing message from extraterrestrial intelligence enables scientists to assemble a never-before known organic molecule!” And the world would be convinced that the signal contained information. The signal demonstrates syntax, semantical content (it turns out to mean the instructions for creating DNA), and thus the intentionality is implied! The obvious inference is that some alien intelligence is intentionally trying to convey instructions to create a DNA molecule.
The inference is so obvious, that it is impossible to imagine this scenario playing out at SETI without scientists broadcasting their success at discovering an alien intelligence.
Just as the computer example illustrates, if you can detect a “program,” you are detecting intelligence insofar as you are detecting the intention to form, convey, and manipulate information. And as the SETI example illustrates, discovering such a “program” in any otherwise random stream of signals would be tantamount to discovering extraterrestrial intelligence.
SETI is not looking for anything like the unintentional “pattern” of growth in either of the pictures of our island example from last week! SETI is looking for something like the code for DNA to appear in signals it collects!
Just as we correctly see intentionality in the existence of a computer and its programming, we see intentionality in both a living cell and in the DNA that is akin to its “programming.”
If scientists “received” the DNA code via a SETI stream, they would immediately recognize it as both information and as intentional. For some strange reason, when scientists look at the living cell and see the DNA code there, they acknowledge that it is information, but they also deny any intentionality in it. By now, the schizophrenia of this perspective should be quite apparent.
But it is not just in DNA that we see genuine information and intentionality. We as fully-formed beings are laden with it. Not only is every cell of our bodies filled with information, but we ourselves are information-processing, information-creating, and very, very intentional creatures!
The empiricist evolutionist has an impossible row to hoe to explain how non-information gets transformed into information without a shred of intentionality in the picture. And the problem is even worse than that, because we are not only talking about the information we can see in a cell, for example. We are talking about the fact that we know ourselves to be intentional creatures; so the question is pressing how all of this supposed unintentionality produced intentionality.
Yes, the cell itself is mindless and unintentional, its processes obeying the instructions it was given, just as a computer mindlessly obeys the instructions it is given. But the fact of the matter is that the computer is manipulating what we know to be information according to intentional rules that we know were coded into it. And DNA alone demonstrates information in the strong sense, not to mention that some of the creatures coded-for by DNA are informational creatures in the strong sense (namely: human beings). All of this clearly rule-governed information-processing implies intentionality and intelligence. Indeed, SETI would instantly recognize it as such if it was presumed to come from some “alien” source!
So, it is obvious that information exists, and it exists in some astounding forms. The very existence of information in any form, particularly forms that we can recognize as such, proves that there are pre-established rules and semantics that must have been intentionally put into place prior to the information itself.
Put another way, “Information” implies both a “reader” and a “writer.” And the “writer” necessarily has logical priority. The rules and semantics had to be in place prior to the information that was created and conveyed according to those rules and semantics.
Information is: Semantical content intentionally formed and conveyed according to syntactical rules.
We see countless examples of “states of affairs” possessing all of these properties. We see it in DNA, and we see it in ourselves qua intentional, information-processing entities. The existence of information demands an explanation of how it is possible.
The empirical/nominal explanation founders, as it necessarily must, because information’s existence depends upon non-empirical entities.
Syntax is an abstraction. We “see” it when we already know what to look for. Thus, the “rules” of syntax already exist in us prior to us “seeing” it in the world of our experiences. Kant provided a compelling proof of what syntax really is, and it is necessarily non-empirical.
Semantical content is an abstraction. As we will further discuss next week, it depends upon the existence of abstract objects called propositions. These entities are not empirically-accessible. Indeed, like syntax, our experiences themselves (and our ability to “think” about those experiences) depends upon these abstract entities. They will forever defy empirical analysis, and Carnap’s distant hope that we would eventually come to have a nominalist account of meaning without appeal to such entities is now dashed.
Intentionality is an abstraction. We know that we have it, and we know that information implies it. But we don’t understand it, even though we know it to be a necessary condition of intelligence and information. Empiricist scientists try to deny it everywhere, saying that life in all its forms is blind, purposeless, and unintentional. Yet, here it is.
So, I believe that the very existence of information will forever defy an empirical and evolutionary account of it. And I would note that none of the references I am employing in these pages and references come from “religionists” who have a religious ax to grind. These are all atheistic scientists and philosophers who are simply being intellectually honest about the genuine problems inherent in a purely empirical/materialistic account of reality.
Finally, let me remind the reader that I do not abide the inference: “We can’t empirically/materialistically ________; therefore God!” Not only is that a flagrant appeal to ignorance in most contexts, but even in contexts in which we can demonstrate that there will in principle be no empirical/materialistic explanation, the “leap to God” is fraught with peril!
As just one cautionary point, the term “God” does not necessarily (or even likely) mean “The Judeo/Christian God as ‘revealed’ by some particular interpretation of the Torah/Bible.” There are so vastly many inferential steps needed to “establish” an “intelligent designer” that the prospect is daunting. And even then, you are logical light-years away from establishing anything like the Judeo/Christian God. I want to emphasize that even if everything I say in this seminar is absolutely true and proved to be true (a bar we are far from getting over!), all we have accomplished is to open the logical space for some sort of designer beyond our ken. That’s a significant step, but it doesn’t get you “God!”
For Further Reference
These book-snippets are very useful in themselves, but you can also purchase the entire books, which I highly recommend.
Metaphysics, Mathematics, and Meaning — By Nathan Salmon, professor of philosophy of language at UC Santa Barbara, this book defends Church’s Translation Argument against the few remaining objections, stating that the objections are either easily answered or based upon misunderstandings of Church’s argument.
I am partial to Church’s Translation Argument, having studied it from Dr. Salmon himself, who studied directly under Church himself. To call Salmon a “luminary” in philosophy of language is an understatement! The linked snippet itself is sufficient to show Salmon’s command of the subject, as well as demonstrate the basis of some confusions critics have had about the subtleties of Church’s argument.
If you read none of the other supplementary materials, I urge you to read this linked text on Salmon’s work on Church’s Translation Argument. Church proved that propositions exist, and Salmon explicates and defends the proof.
Kant and the Empiricists: Understanding Understanding — This book by Wayne Waxmen of the University of Colorado contrasts Kant with the famous empiricists. Particularly his section on triangles is very penetrating.
Challenges to Empiricism — Herein, Harold Morick rigorously considers the empiricism of Carnap and other modern empiricists. This snippet addresses the paradox of analyticity and concludes that empiricism has no adequate approach, much less account, of how to respond to it.
Decline and Obsolescence of Logical Empiricism: Carnap Vs. Quine and the Critics — The title says all, and the excerpt argues compellingly that the semantical tradition embodied in Logical Positivism was doomed and that we now understand the roots of its demise.