Virtualization and Understanding Consciousness

Virtualization is a huge topic, worthy of a dozen posts, but let me sketch it out here in the hope of making some details clear later.

ARCHITECTURE

I'm no fan of using computers as models of "mind". Our minds are not computers, nor do I expect that the AI dream of making a mind out of computer chips will happen any time soon. However, it is clear that both computers and minds process "information" in some broad sense of the term and we should not be surprised to find that designers of computers and "mother nature", the "designer" of our minds, have face similar problems and come up with similar solutions.

COMPUTER ARCHITECTURE

If you know a bit about computers, you know a bit about "virtualization" in computer design. The Central Processing Unit (CPU) of a computer manipulates a very small amount of information very quickly. This information is swapped in and out of the CPU from "on board" memory - Random Access Memory or RAM. RAM has always been relatively expensive, so computers have been swapping RAM in and out of memory to some version of the "hard disk" since the 1960's. So there is a kind of "bucket brigade" or hierarchy of storage, that makes huge amounts of information accessible to the CPU in a fairly painless way. A lot of this swapping takes place more or less invisibly or "under the hood" and does not concern the computer programmer.

THE CLOUD

Most recently, with the advent of languages like JAVA and object-oriented design (which cares not where the object "lives") location of the information (data or program) has become irrelevant as chunks are swapped in and out of the "cloud" and between "hosts" with all the "plumbing" functioning automatically. We wind up with our CPU's manipulating information from all over the world, selected automatically from a vast amount of information made available on hosts all over the planet.

That's virtualization.

VIRTUAL MINDS

Something similar goes on in our minds.

What we call "working memory" is very similar to what we call "attention" or "consciousness". It is what it feels like when a portion of our brain activates a very small subset of what the brain "knows". This process is powerful and mysterious - quite unlike what goes on in a computer - but it works in a similar way to address similar problems that we see in computer design. For example, we can "pay attention" to about half a dozen "things" at once. We can switch things in and out of attention a few times per second, producing the continuing experience we call "consciousness" - the flow of "things" through our CPU of working memory. What's more a "thing" can be as simple as pressing the "E" key on a keyboard of the idea of representative democracy. Our minds are perfectly capable of flipping around "things" very quickly as we type away furiously on the keyboard thinking and writing about political philosophy while trying to remember how to spell "philosophy". Almost magically, "ideas" that we have packed away in our brains pop in and out by means of their connections to ideas held in attention a fraction of a second earlier.

LANGUAGE AS THE GREAT VIRTUALIZER

Language is a great virtualizer.  A word like "Canada" can be used to reference a vast amount of information stored away in the brain awaiting unpacking, resorting and repacking. What's more, such a word can be transmitted via speech to another brain, whereupon it proceeds to unpack "meaning" in the second brain. The equivalent of the "cloud" is what we call "culture". Those plugged in to "culture" have access to a vast ocean of information shared among members of that culture. Most of us belong to more than one culture - for example there is an international culture of Scientists and Intellectuals for whom language differences are a mere inconvenience to sharing of ideas.

Without getting too spooky or mystical, I think it's not entirely wrong to see the cultural equivalent of the "cloud" as a group mind - something that is alive and purposeful and not a mere information store.

BEES

Consider the bees ...

An individual bee is an incredible product of evolutionary engineering. Its ability for controlled flight would be the envy of a veteran helicopter pilot. With a microscopic brain, it is capable of navigation and able to communicate the location of promising flower beds to its hive mates.

But it doesn't know it's a bee. 

Everything the bee does is in the service of the hive. It lives and dies to ensure that the hive will survive through the winter. The hive as a whole seems to "know what it's doing", yet the hive has no mind of its own. Hive "behavour" and hive "knowledge" somehow emerge from the collective, individual behavour of bees. I won't go wandering off into abstract speculation to speculate on how this happens. My point is simply that it does happen. The hive mind is distributed among the minds of tens of thousands of bees, each with a lifetime measured in months at most, each with a remarkable ability to sense the world and act in it, but each totally "unaware" of what it is doing.

It is also important to note that the "hive mind" has been shaped (and is being shaped) over millions of years in a wonderful dance with the flowers, who "use" the bees to reproduce sexually and thereby maximize the chance of evolutionary variation. And of course, a rose doesn't know it's a rose.

DO WE KNOW WE ARE HUMAN?

So we are tempted to ask, do we know we are human beings? Is there a level of reality in human history that is forever beyond our ability to understand, just as a bee will never understand the hive it dies for and a rose doesn't know it smells sweet? This is hardly an original question. Religions pretend to answer it by somehow uncovering the secret thoughts of "God", along with the heavenly "big picture". I don't think I'm alone in thinking these efforts result in little more than childish fairy tales, much smaller than the phenomena they attempt to "explain".

NOT SURE WHAT IT IS, BUT HOW DOES IT WORK?

I think we can step back from telling ourselves mystical stories and learn a lot simply by looking at the architecture. There is a hierarchy of information processing that goes all the way from individual "attention" to cultural knowledge. Indeed, since cultural knowledge relies at least in part on experience of the real world, we can regard physical reality as our ultimate "cloud" storage, just as the story of the bees is incomplete without accounting for the role of the flowers. However, like bees, we are restricted by this architecture from experiencing any of it except through a tiny window of "working memory" - less than a dozen "chunks" at a time. Language itself is our "operating system", providing reasonably efficient ways to "download" chunks from our "cloud", process them and store them back in the form of new "mash ups" of old ideas. Human language plays a similar role in our minds as Java plays in the computer you are looking at now. Ideas are popping into your head through words and it makes no difference where these ideas came from. The only requirement is that both the author (me) and the reader (you) are both human - running the basic operating system, capable of turning ideas into words and vice versa.

Of course, this insight applies far beyond the realm of language and "words", but that's a topic for another essay ...

Consciousness and the Dragon

It may be said that whether or not a corporation is "alive" is a matter of definition. I find it quite easy to make the necessary adjustments. But we are still left with our doubts about the analogy. We expect a living thing to be "like" us - especially to have something we call consciousness.

Consciousness is a slippery idea. For most of us, it refers to the experience we have of inhabiting the world. It is subtly connected to the phenomenon of attention. A vast amount of my behaviour is obviously controlled by my brain but not in a conscious way (I don't need to think about how to walk down the street, drive a car or decipher letter patterns while I read). Consciousness (attention) is about what I do need to "think about".

This distinction is made in "The Ravenous Brain", a very readable source for those looking for an update on the current status of research into how the brain works. In summary, the function of "attention" or "working memory" is to deal with anything unusual by either labelling it as something previously experienced (and moving on) or by "chunking" the new experience by categorizing and pattern recognition, then "filing away" the new "chunks" for future reference. As someone with computer background, I think of these "chunks" as active "apps" - objects that allow for extensive generalization. We used to call these "subroutines", but now the best word might be "types". In any case, the result is that our "chunks" allow us to "think" quite extensively without our interactions with the world entering into "working memory". They are automatic but definitely not "static".

I contend that it's easy to characterize a corporation as being alive and "thinking" in an unconscious manner. This involves running "on automatic", following set rules that admit to no exception. Considered as a live thing, the corporation is very stupid. When we look for an analogue to human attention and consciousness in a corporation, we must admit that it comes from the human beings that control the corporation - especially those in the corporate suite. These people actually pay attention to circumstances and draft new rules to be buried in the corporate mind in the same way that a human learns to ride a bike, then never needs to think about it again. This is especially evident in the case of the military, where some level of creativity is permitted at each level, while the entire organization is designed to operate as a machine with the generals dealing only with exceptions and unusual aspects of the particular situation. The extreme case comes when a human being (President of the United States) decides to initiate a nuclear attack, which will happen entirely automatically about four minutes after his decision. The fact that is entirely automatic (intelligent but automatic) is an essential element of the terror it is designed to provoke in the enemy.

The creative accounting procedures used by ENRON before it blew up were the conscious creation of identifiable human beings, but the "mind" of ENRON extended far from the deliberations of these few individuals. For the vast number of ENRON employees, it was simply a matter of going to work in the office and working out the details. Aside from a few perversions of basic business practices, ENRON was just another corporation.