« Back to Table of Content || Next »

Chapter I: Uniqueness of the Human Mind >> Language as an indicator of mental synthesis

Language as an indicator of mental synthesis

Owing to the fundamentally subjective nature of the process of mental synthesis, we have to rely on indirect evidence of mental synthesis capabilities in animals. Specifically, we will look into two lines of evidence: (1) expression of mental synthesis through language and (2) the ability of animals to solve complex problems.

Human language reflects the process of mental synthesis

The human mind has a unique ability to synthesize new images from those stored in memory. Not surprisingly, human language has evolved to allow one human to synthesize images in another human’s mind. In the example of “a cup on top of a keyboard” that we discussed above, your mind remembered a cup, remembered a keyboard, and then synthesized a new mental image of a cup standing on a keyboard. I used language to synthesize a new image of “a cup on top of a keyboard” in your mind. I can now add a flower: “a flower in a cup on top of a keyboard”. Therefore, I was able to use language to direct synthesis of a new image in your mind from images that you already had in your memory.

In fact, humans primarily use language to describe something that the listener has never seen. That is, humans use language as a tool for synthesis of new images in the listener’s mind. Consequently, language can be used as a window into the mind. In fact, just by listening to a human conversation, an observer who understands the language can conclude that humans are capable of the process of mental synthesis.

Language reflects the communication needs of the mind

It is thought that a language reflects the communication needs of any group of humans. For example, the English language has only a few words that describe snow. Some Eskimo languages have significantly more words describing snow. The anthropologist Franz Boas in “The Handbook of North American Indians” (1911) identifies the following words used by Eskimo to describe snow: aput ("snow on the ground"), qana ("falling snow"), piqsirpoq ("drifting snow"), and qimuqsuq ("snowdrift"), as well as "akerolak" ("newly drifted snow"), "perksertok" ("drifted snow"), "pokatok" ("grainy, salt-like snow"), "mauyak" ("soft snow"), "patu ("frost"), "minu" ("light frost"), "illuyak" ("frost on a window"), and, my favorite, "ayak" ("snow on your boots"). It is thought that Eskimo have more words for snow because they have to deal with snow more often.

Portugal has a well developed coffee culture. Linguist Marina Khabensky identified these Portuguese words that describe coffee with milk: “meia de leite normal” (prepared with coffee that was already once used for brewing, served in a medium cup), “meia de leita directa” (prepared with fresh coffee, served in a medium cup), “pingo normal” (prepared with coffee that was already once used for brewing, served in an small espresso cup); “pingo directo” (prepared with fresh coffee, served in an small espresso cup), “galão normal” (prepared with coffee that was already once used for brewing, served in a large glass), “galão directo” (prepared with fresh coffee, served in a large glass), etc.

Even subgroups of people within one language tend to expand their vocabulary as needed. Skiers in USA use the following words to describe the quality of the slope: champagne powder (the dry, crystalline powder snow), crud (soft and mushy snow), freshies (new fallen snow), chunder (choppy, wet snow), concrete (icy conditions), corn (semi-frozen spring snow), boiler plate (super hard icy surface), cement (wet, heavy snow), formica (icy conditions that remain after the powder has been skied off), frozen smoke (very dry powder snow), hero snow (optimum conditions), mashed potatoes (sloppy, ungroomed snow), piste (groomed snow), pow (powder snow), scrambled eggs (sloppy, ungroomed snow), etc.

By looking into language, we can understand the unique communication needs of a particular group of people (or animals). Accordingly scientists have always been interested in studying the language used by animals in the wild.

Animal use of language in the wild

Animals in the wild often use language of signs and vocalizations for communication.

Bees. Many species of bees use dance to communicate the direction and distance to a food source. Karl von Frisch correlated the runs and turns of the dance to the distance and direction of the food source from the hive. The orientation of the dance correlates to the relative position of the sun to the food source, and the length of the waggle portion of the run is correlated to the distance from the hive. Karl von Frisch was awarded the Nobel Prize in Medicine in 1973 for his proof of the language and dialects of bees.

Birds. Bird songs play a role in mating as well as identification of the singing bird. Experiments suggest that the quality of a bird’s singing may be a good indicator of the bird’s health as diseases can degrade song quality. Individual birds may be sensitive enough to identify each other through their calls.

Whales. Two groups of whales, the Humpback Whale and the Blue Whale, produce repetitious sounds at varying frequencies known as whale songs. Humpbacks make a sound called the feeding call (10 seconds long sounds of near constant frequency). Since Humpbacks generally feed in groups, it is likely that feeding calls help whales assemble in groups. Male Humpbacks are also known to perform songs during the mating season, probably for attracting a mating partner.

Dolphins. Dolphins emit two very distinct kinds of acoustic signals, which are called whistles and clicks. There is strong evidence that some specific whistles, called signature whistles, are used by dolphins to identify and call each other. Dolphins develop their own unique signature-whistle early in life. It was documented that dolphins both generate other dolphins' signature whistles, and their own.

Vervet Monkeys. As mentioned above, the Vervet monkeys of East Africa sound different warnings for different threats: a leopard warning that sends vervet monkeys climbing into trees, an eagle warning that makes them hide under bushes, a snake warning that makes monkeys stand on their hind legs, a baboon warning that sends monkeys rushing to treed areas.

Chimpanzees. As mentioned above, chimpanzees in the wild have a number of words. They have words for enjoyment, excitement, puzzlement, annoyance, distress, anger, etc. If one chimpanzee sees a snake, it makes a low, rumbling noise, signaling all the other chimpanzees to climb into nearby trees. When a food source is located, the chimps can vocalize with loud calls to inform other chimps that food has been found. When seeing a leopard, chimpanzees give loud alarm calls described as ‘waa barks’ by Jane Goodall who studied chimpanzee in the wild for over 45 years. Chimps can explain this simple information to other chimps but they cannot express the idea that the leopard is approaching from the left or from the direction of the sun.

Thus animals in the wild use signs and acoustic calls to communicate information. However this information is limited to a description of feelings, danger codes, and, in the case of bees, distance to the food source. I surmise that the language of chimps only evolved to support what the chimps’ mind is able to imagine. If so, since a chimps’ language at best enables chimps to describe their emotions and a few predators to other chimps, we have to conclude that a chimps’ mind can only imagine emotions and these predators. Since the chimps’ language cannot describe to another chimp a never-before-seen object, we can infer that chimps have no use for such language. In other words, we must conclude that chimps cannot synthesize a new, never-before-seen image in their mind.

It would not benefit a chimp’s survival to have separate words for predators (a leopard, a snake, a lion) and elements of the environment (a stone, a tree, a river) if chimps were not able to imagine the predator in the context of the specific environment in order to infer the exact location of the predator. Suppose chimps understood the word “snake” and multiple words that describe trees and stones: “tall tree”, “short tree”, “green tree”, “tree with big leaves”, “big stone”, “white stone”, etc. Now suppose that snake is approaching from the direction of the tall tree located near the big stone. What would be the use of the call “a snake - the tall tree - the big stone”? The chimps cannot use mental synthesis to synthesize an image of “the tall tree near the big stone”. Therefore, they are not able to infer the location of the snake, and thus react by running in a direction away from that tree. The effect of the words “a snake [is approaching from the direction of] the tall tree [near] the big stone” would be exactly the same as the effect of the single word “snake” - namely, all chimpanzees running for their lives in different directions looking for a nearby tree to climb on. In fact, chimps do have a warning call for snakes - a low, rumbling noise (chimps react to this call by climbing on nearby trees) and do not have words that describe elements of the environment. They do not have words or signs for trees and stones not because they are limited in the ability to generate such signs (as noted below, they can learn hundreds of signs) but because they do not have much use for them. Based on the content of chimp language, we have to conclude that, at least in the wild, chimps do not develop the capability of mental synthesis.

In the following discussion we will need to be able to refer separately to the animal language of signs and codes and to complex human language. However in the discussion of human evolution we will no longer be able to use descriptions such as “animal language” or “human language”. I was looking for a one word descriptor that identifies the difference between human and animal languages. I have considered and decided against “symbolic language“ since both human and animal languages could theoretically be deemed symbolic. Remember that Vervet monkeys’ warnings of a leopard, a snake, an eagle, and a baboon could be considered symbols for those predators. I have considered and decided against “syntactic language”, since both languages (at least theoretically) could be deemed syntactic. Syntax is the study of the rules that govern sentence structure. Since the sequence of animals’ calls may be in some cases important for communication, syntax may not be a straight forward separator.

Since I was not able to find a single existing descriptor that identifies the difference between human and animal languages, I felt that I needed to introduce a new descriptor. Human language evolved to allow the speaker to synthesize new mental images in the listener’s mind. Therefore I propose to call it a synthesizing language.

All known animal languages are not suitable for synthesis of new mental images in other animals’ mind. Accordingly, I propose to call them non-synthesizing languages. The non-synthesizing language allows animals to communicate emotions, warnings, maybe even memories of objects or events. However, the non-synthesizing language is not intended for synthesis of images that the listener has never seen before.

Modern humans have a synthesizing language. Chimpanzees and other animals in the wild have a non-synthesizing language. However, maybe animals can be taught a synthesizing language in a lab. Then, by talking to an animal, we could infer its capacity for mental synthesis. The next section summarizes attempts to teach animals a synthesizing language.

Can animals be taught a synthesizing language?

Parrots. Parrots are famous for their ability to mimic human language. I am always reminded of this fact when visiting Manhattan. I usually stay at the house of my friends who own an African Grey Parrot named Kaipo. Kaipo likes ice cream and apples. Whenever he feels like eating dessert, Kaipo loudly and clearly says “Kaipo wants ice cream” or “Kaipo wants an apple”.

Another African Grey Parrot, Alex (acronym for Avian Learning EXperiment), learned about 150 words and was able to answer a number of simple questions about objects. Trained by animal psychologist Irene Pepperberg, Alex appeared to have an understanding of what he said. For example, Alex could identify fifty different objects, was able to distinguish seven colors and five shapes, several materials (wood, plastic, metal and paper). He also understood the concepts of "bigger", "smaller", "same", and "different", and he was learning "over" and "under". When Alex was asked about the difference between two objects, he was able to answer correctly. When there was no difference between two objects, Alex said “none.” Alex was able to recognize quantities up to six. In 2005, Dr. Pepperberg reported that Alex understood the concept of zero.

When Alex was tired of being tested, he would say “I’m wanna go back,” and if the researcher displayed annoyance, Alex tried to defuse it with the phrase, “I’m sorry.” If he said “wanna banana”, but was offered a nut instead, he stared in silence, asked for the banana again, or took the nut and threw it at the researcher.

Project Washoe. Non-human primates lack vocal cords. For that reason, the number of different vocalizations is greatly limited. To circumvent this problem sign language and lexigrams (symbols that represents words) are used to teach language to primates. Washoe was the first chimpanzee to learn American Sign Language. Washoe could reliably use about 250 signs. Washoe also displayed the ability to combine signs in novel and meaningful ways. For example, she referred to the refrigerator as “open food drink”, even though the scientists always called it a “cold box”.

Alan and Beatrix Gardner used conditioning to teach Washoe sign language. After several years Washoe could learn new sign language gestures without conditioning, simply by observing humans signing amongst themselves. For example, the scientists signed "toothbrush" to each other while they brushed their teeth near Washoe. On a later occasion Washoe reacted to the sight of a toothbrush by spontaneously producing the correct sign, thereby showing that she had in fact previously learned the sign.

Koko: A Talking Gorilla. Koko is a lowland gorilla. She has been trained since the age of one by Dr. Francine Patterson. By the time Koko was 10, Dr. Patterson had assessed Koko's vocabulary at over 1,000 signs, which places her among the most proficient non-human users of language. She can also understand approximately 2,000 words of spoken English. Dr. Patterson has documented Koko inventing new signs never taught to Koko: Koko combined the words "finger" and "bracelet" to describe a ring; invented "drink-fruit" to describe a melon, and invented "water-bird" to describe a swan. In August 2004, Koko communicated that she had a toothache. Several documentaries have been made about Koko, including “Koko: A Talking Gorilla” (1977).

Project Nim. Dr. Herbert S. Terrace led a study of a chimpanzee, Nim Chimpsky, at Columbia University. Nim Chimpsky was given his name after Dr. Noam Chomsky, the theorist of human language structure and grammar. Project Nim was an attempt to go further than Project Washoe. Project Nim was conceived as a challenge to Chomsky's thesis that only humans can grasp a complex language. Nim lived 24 hours a day with his human family from birth and learned about 125 signs. He was able to “say” short sentences like “Yogurt Nim eat”, “Play me Nim”, “Tickle me Nim”. After many years of training Dr. Terrace concluded that Nim hadn't acquired “real language”.

A “real language” is expected to have a complex syntax. In human language, switching words in a sentence, like “cat ate mouse” to “mouse ate cat”, changes the meaning of a sentence. In this book we defined this type of language as a synthesizing language. The sentence “cat ate mouse” synthesizes an image of a cat biting on a mouse in a listener’s mind. The sentence “mouse ate cat” synthesizes an image of a big mouse biting a smaller cat in a listener’s mind.

Nim and other animals have never learned to understand the changes in the meaning of the sentences when the order of words was changed as in the example above. They learned to repeat signs, previously taught to them by researchers, in appropriate contexts. Their use of language was strictly pragmatic, as a means of obtaining an outcome. In other words, their language was more like using codes. That is the type of language that we refer to as the non-synthesizing language.

The biggest difference between the synthesizing and non-synthesizing languages is that the latter is not intended to generate a never-before-seen image in the mind of a listener. Sentences like “Kaipo wants ice cream” and “Yogurt Nim eat” are not intended to create a new image in the listener’s mind. These are codes that intend to trigger a preexisting reaction in a listener.

In fact, sign communication may emerge spontaneously in animals interacting with humans. The most obvious examples include a dog that shows its owner the house entrance door as a request for taking it for a walk, and a cat that takes its owner to the refrigerator as a request for food. Smarter animals can learn to use hundreds of words appropriately, as we have seen in the experiments with talking birds, gorillas, and chimpanzees. I agree with many scientists who, after multi-year efforts to teach animals words and signs, have concluded that all these words and signs were used as codes. Even when animals were able to explain complex instructions, it was just a longer sequence of codes. I have never seen an experiment from which I could deduce that an animal first synthesized a new image in its mind and then generated an instruction. Even when Koko created new words for objects that she didn’t know the names for, she did not need to use the process of mental synthesis.

Therefore, we have to conclude that despite years of training, no animal was ever able to use a synthesizing language. In all likelihood, the reason for this, is a lack of mental capacity to synthesize new mental images. Animals cannot imagine something they have never seen. I suppose that animals’ visual system can be likened to a video camera. Animals can only record what they see. Of course they can watch the recorded movie later from their memory, but they cannot change the sequence of frames or rearrange the actors.