There�s a Diamond in My Pocket

Heft arose from his deep sleep. As he sat up in the cot he turned to his brother still sleeping and said with enthusiasm, �Arise Pious, it is early morn, and we must tap the beer keg for it is fathers birthday today! �

  Pious rumbled from his slumber and arose from his cot without a word. Moving across the dirt floor, Pious reached the closet where the family kept their cloths-dress, grabbed a vest from the coat rack and headed out through the wooden door, �Wait for me! � Heft exclaimed as he shuffled his massive body following in quick pursuit.

  The two brothers arrived at the cellar door a few yards from the cottage. Heft grabbed the massive wooden mallet and swung it over his shoulder-Pious already had the iron tap in hand. Opening the door they entered the dusty cobweb hole that housed the beer keg and some wine bottles.

  The two nomads in the cavern moved silently to their destination as driven by some unknown force. Pious held the iron tap over the corked hole in the beer keg and muttered, �I don�t know who is going to be happier today-me or father.� Heft acknowledged his brothers statement by drawing back the massive wooden mallet and hitting the iron tap with such force as to drive the cork into the keg and replace it with the iron tap knocking Pious to he floor.

  Pious looked up from the dirt floor at Heft and knew they had been successful by the large grin on his brother�s face. A glimmer of light caught Pious�s eye in the dark cavern. Hay look Heft, the light from the cellar door is making something sparkle.�

  �What is it?� Heft asked with curiosity.

  Pious reached over from his position on the dusty floor and grabbed the sparkling object. �Why �it�s a diamond. There�s a diamond in my pocket. I thought I saw something fall out of it as I was knocked down!�

�Oh no!� cried Heft. �Your wearing fathers vest!�

  In 1950, Alan Turing published his now famous paper, Computing Machinery and Intelligence. The basic question of the paper is �can machines think?� The answer to this question is yes. However, a condition set forth by Turing must be met. It is known as the Turing Test. The test is based on a charade game people played in which a man and woman would sit behind screens and a third person would try to guess through there conversation who was the man and who was the woman. Using this same scenario, Turing devised as system in which a computer and person were behind the screen and the third would try to tell which was the computer and which was the person. If the third person could not tell which was the person and which was the computer-then the computer could be considered to think. This is the Turing Test. The artificial intelligence (AI) community has used this as a standard to test for thinking machines. Conversely, it has come into question as whether it is valid or not.

The most dominant feature of the Turing Test is the use of language. Therefore it would seem that the effective use of language is not only the criteria for thinking, but also the definition. This is not the first implication throughout history that defines thinking as to be connected with language. The philosopher Rene Descartes argued in 1642 that machines could pass as animals; however, they could never pass for humans. This assertion is based on the machines ability to use language and to reason. Oxford Dictionary defines the ability to reason as �the ability to think and draw conclusions.� This definition is incorrect? It is known that chimpanzees think, that elephants think, and a number of species of animal�s think. I argue this fact and others such as philosopher Jack Copland and other professionals in various fields. Thus, language is not a defining feature of thinking, only a measure of it. This Turing test will never identify thinking if used in this context. In other words, the Turing test will never work unless it is put into its proper context. And this is precisely my objection. A good test is thought about all wrong. In addition, is the disregard of the use of language as a proper meter stick for finding intelligent machines? Lets look at the following first.

  In the case of the Turing Test, it is the case of mistaken identity. Just as if some one in the family has the same type and color of shirt or vest and the same statue, it wouldn�t be to hard to mistake it for your own if you were in a hurry out the door. The same is with language. It is the mistaken vest of thinking. If chimpanzees can think and not speak a language suitable to pass the Turing Test, then language does not equate to thinking. Before we look at language further, let us turn our attention to a definition of thinking. If chimpanzees think and humans think, then both species think. Also, if thinking is describable and identifiable without the use of language, then thinking is something tangible. Thus, chimpanzees think and humans think, if and only if thinking is something tangible. Here I have defined thinking as something tangible. Some of its properties stand out. It has a regular structure. It is both recognizable and identifiable. It has regular and particular components. An analogy is the properties of the atomic structure of a diamond. It has a regular structure, and the particular components of carbon construction with a high level of hardness. Moreover, any person on earth would recognize a nicely cut diamond. Consider a modified Turing Test or more appropriately a new contest.

Language is a measure to identify thinking. The language itself is a construct between individuals of a group, however indiscernible as it may be. That is, the language may be so alien to Homo sapiens; it may only be identified by its structure. For example, prairie dogs circle around dieing family members and pay their last respects. This is known by behavioral analysis. Thus, the strange sounds made by prairie dogs are a form of communication, i.e. language. Let�s look at the following example to clarify why. Consider the following statement. All odd numbers that are less than 29 or all divisors of 48 are Glacks. Consider this statement to be true and equivalent to our diamond. In other words, these conditions are met if and only if there is a thinking machine. Now let�s test for a thinking machine. Test one: 16 are a Glacks. This is a true statement due to 16 is a divisor of 48. We have found a thinking machine. Test two: All Glack numbers less than 15 are odd. This statement is a false-2 divide 48. We have found a machine that cannot think. Test three: No odd number greater than 27 are Glacks. In this statement we can�t tell. The statement only says that odds less than 29 are Glacks and sys nothing about odds greater than 27, 29 etc. The statement makes no guarantee of this. This means we are also in the dark about whether our machine is or is not thinking. This is the crux of my argument.

 In the first test we used a method that yielded a true result that made thinking discernable, identifiable, and thus tangible. In the second test we used the method, which yielded a false result, and produced a tangible. However, in the third test we just can�t tell. There is nothing describable, identifiable and thus nothing tangible. In the case of species that populate the Earth, from insects to humans, you can find all three true or false, or can�t tell statements. For insects and worms, they�re about their busy work without much regard to the complexity of their behavior, i.e. not thinking or just can�t tell. In the case of Chimpanzees, Jane Goodall sates, �chimps are very quick to notice and interpret the eye movement of their fellows, and Goliath (the chimp of focus) would possibly, therefore, have seen the fruit himself. And so Figan (The interacting chimp) had not only refrained from instantly gratifying his desire, but had also, gone away so that he could not give the game away by looking at the banana.� Here tracking the banana and interpreting eye movements indicate acts of thinking-something describable and identifiable-thus tangible. So we can say chimpanzees think. For humans it�s a given. But in the case of thinking machines and a worm, you just can�t tell unless there is a measure. For the worm, it must be shown that there is behavior that leads to communication among individuals of the group. For the thinking machine, the language may be that of human language or that of 1�s and 0�s. Otherwise the machine may look like an undersized tool shed. To avoid this, some have suggested giving machines sense organs-maybe give them eyes. This just adds to the complexity of the problem. However, behavior such as the chimps making contact with their eyes, computers can make contact with binary language. Both are adequate as measuring devices, i.e. metrics. This is a unique opportunity to remove machine from the �can�t tell� class, simply because there is an effective metric. We can now look for the diamond.  

In looking for a thinking machine, language can reflect the thinking process. AI already has some elements in their machines that reflect thinking but don�t quite measure up to thinking. The machines lack understanding even though they have complex knowledge representation of schemes. Knowledge representation schemes such as forward and backward reasoning, reasoning with uncertainty, non-monotonic reasoning, probabilistic reasoning, fuzzy reasoning, reasoning by analogy, and case-based reasoning. Many of the knowledge representation schemes are self-descriptive by their names. Though they are separate entities now, later they will most likely be called reasoning modalities. Parts of a larger knowledge representation scheme. Reasoning also comes in �flavors,� induction, abduction and deduction. Induction is something you expect, abduction is reasoning back from something, and deduction is mechanical-what happens if I set this tool here. All these things can be found in AI machines, which are shared by humans, and cannot be tested for by the use of language. But whether a machine is thinking by fooling a person into thinking they are talking to another person is a facade. It has been done. Is this a measure of thinking as stated by the Turing Test? No. Instead, language needs to be used to test for understanding, the modalities I spoke of earlier and possibly some other yet unidentified properties as part of a larger knowledge representation scheme. The thinking machine must be able to make critical connections between applications and apply their knowledge scheme to situational platforms. Not the mechanical application of a knowledge scheme, guided by a human to situational platforms.   

�What do you have there Lorn, � Pluvial said in a breathless tone.

�Well let me tell you about who I have here first,� replied Lorn

�Have you gone mad?� Pluvial scorned.

I purchased a cottage up in the mountains last summer. In fixing up the place I had to put a composite foundation under the cottage to replace the old cement. When I lifted the cottage off the old foundation with the Graviton, I found this diamond in the far right corner of the house near the cellar.

The story goes that two brothers, Heft and Pious, had stumbled on to their father�s only possession of wealth. It�s a diamond the father found while clearing the property for the original cottage. He took it to town and had a local cutter shape the diamond. I think Pluvial they became friends after that. Unknown to his sons and the towns people, he carried the diamond in his vest pocket.

Apparently, on their father�s birthday, they got him so drunk he couldn�t remember the day. Later, when he realized the diamond was missing, he approached the two brothers and told them the story of the diamond. He then asked the two if they had seen the diamond around or near the cottage.  The two brothers convinced the father that he must have lost it in his drunken stupor. For whatever reason, they hid it in the corner of the house. They never sold it for money; they were probably terrified someone would tell the father.

�It�s a wonder that the thing was not lost eternally in the five renovations that have been made to that cottage over the centuries.� Pluvial commented.

�I don�t think it wanted to be lost eternally. And I don�t believe the brothers could sell this diamond. I believe it lives there at the cottage and will return home this summer when I go on vacation.� Lorn returned.

  Pluvial stood breathless and motionless before the diamond sparkling in Lorn�s hand.

�That�s right, we have to give this diamond the New Turing Test. We can�t use the language setting. The hardware it possesses is quantum mechanical. It�s embedded in the electron cloud of the carbon atoms no more than atto-femtometer in length. Well be using the Schrodinger thinking equation today.