REMEMBERING
PAPUA NEW GUINEA
SPOKEN SPEECH and the INVENTION OF WRITING
SPOKEN LANGUAGE and the INVENTION OF WRITING A new book that combines genetic discoveries made in 2003 with information discovered in the study of the development of sign languages to propose a explanataion for the origin of spoken speech in modern human beings. Most of the rest of the book is concerned with the four different societies that seem to have independently "invented" systems for writing speech down and then being able to read it back. That is the groups that made the intellectual leap of producing groups of arbitrary marks on clay, or bone, or stone, or papyrus that could record a poem, or a story, or a song, or a trade transaction, or the payment of a tax, or a hymn of praise for a king. The book considers how they did it, and how our more recent ancestors figured out what they had done. While Sumerian, and Egyptian Hierogyphics, and Maya Hierogyphics are now all dead, Chinese characters are still very much alive. Since this is so, Chinese characters are used, as examples, when considering the use of computers in recording, recognizing, and printing characters and in presenting various examples of machine aided translations between English and Chinese. In addition there are chapters on Sign Languages and Braille and upon the notations used to record music. It was a dark and stormy night, the rain came down in sheets, the wind blew in all directions, the weird rock formations were silhouetted by the flashes of lightning, the horses were tired and one of them was going lame. A break in the clouds allowed the moon to illuminate a billboard with an inscription in large red characters,  How did it come about that these groups of arbitrary marks on a piece of paper or a billboard could lead our imaginations to construct scenes completely different from the one in which we read them? This book uses these groups of letters, signs, or characters assembled into groups, often called words, to explore the efforts of some early humans to make the intellectual leap from spoken speech to what is now called writing. This leap was made, at four different times, in four different parts of the world, and apparently with four different motivations. The four regions developed four completely different looking sets of signs or characters. Only one of the original four scripts has survived in an unbroken chain from its origin until today. Before getting to writing a plausible explanation is provided for the origin of spoken speech in human beings. A historical background is provided for the efforts, during the last 200 years, to decipher the early writing systems. The use of computer based software and hardware is examined in some detail both to help in understanding ancient languages, and in aiding in the present day translation between different languages using different scripts. Throughout the book attention is paid to our pattern recognition abilities and how they aid in producing and understanding these various signs, characters, and scripts. This book was fun to write, I hope that you will find it both enjoyable and informative to read. Kenneth K. Clarke About four years ago I started writing a book on pattern recognition in and by human beings. After I had written chapters on the early writing systems of the Sumerians, Egyptians, Chinese and Maya, on Sign Languages, and Music Notation I got involved in workings within the human cell. To narrow down the discussion, on the last topic, I undertook to explain the path from a segment of DNA known as a Gene to the construction of its associated Protein. At this point I turned back to an introductory chapter that would consider the origin of human speech. Since I consider that writing is a means of recording, and being able to “play back” human speech,without speech, in some form, there would be no writing. While the origin of speech has puzzled men for centuries, and has produced all sorts of far fetched theories, it has only begun to yield rational answers within the last four-five years. It turned out that the topics that I had previously considered provided the a reasonable background to take a new view of this subject and to allow the proposal of a rational solution to it. This chapter combines a brief tour of the history of man, with some facts about molecular biology and heredity, with some facts about mutations, and with a brief discussion of the development of sign languages, to outline this theory of the origin of human speech. More details on all these topics, except the early history of man, are contained in later chapters. A VERY BRIEF OUTLINE OF THE HISTORY OF EARLY HOMINIDS Somewhere on the order of 6 million years ago (mya) the ancestors of the species now known as “man” broke off from the branch of primate family that went on to become present day chimpanzees. This branch is part of a larger group known as mammals that has been around for at least 75 million years. The present day estimates are that since this split there have been about 20 sub-species of hominids, of which Homo sapiens are the only survivors. Since many of these “ancestors” or “cousins” are known only by a single, sometimes incomplete, skull their relationships to each other and to us are not very well understood. While we really do not know for sure, it is not clear that any of these other branches of the family tree ever had the gift of speech To telescope the history lesson we skip over 5.5 million years to about 500 thousand years ago (kya) when a group, now called Heidelberg man , evolved in Africa . Some of this group migrated to Europe where some of them “evolved” into the Neanderthals. Those that stayed in Africa slowly changed so that by about 160 kya a set of “nearly modern” people began to appear. By 120 kya these people had spread out over Africa and slightly into the middle east. Their remains seem to cover the time span from about 120 kya to about 70 kya. Most of these people seem to have perished between 70 kya and 50 kya. Severe climate changes in both northern and southern Africa may be responsible for their demise. Milder climates in the regions of present day Kenya and Ethiopia seemed to have allowed some survivors there. Somewhere in the vicinity of 50 kya at least some of this group went through a fairly rapid transitions into what we now know as Homo sapiens. While we believe that their skulls and skeletons do not differ much from the people of 50 or 100 thousand years earlier, their artifacts and customs did change radically. After a string of digressions we will return to this “bottleneck” and argue that speech, as we know it today, probably originated then and then contributed greatly to these cultural changes. DIGRESSION INTO MOLECULAR BIOLOGY In most non-virus and non-bacterial forms of life each cell contains a nucleus which floats in the liquid cytosol that is contained within the outer walls of the cell. The nucleus also contains the nucleolus Together they contain the bulk of the cell’s copies of the person’s DNA. To provide some size perspective the “typical” human cell might have a diameter of 20 μm and weigh 10 nanograms DNA Our genetic information is contained in our DNA. Chemically a segment of DNA is made up of a five carbon sugar molecule; and two oxygen- phosphate groups, to bind the each sugar to its adjacent neighbors so as to make a chain. In addition one of four different nitrogenous bases is also attached to each sugar. The four bases are ADENINE (A), CYTOSINE (C), GUANINE (G), OR THYMINE (T). Since the sugar molecules are not symmetrical, the DNA chains have a definite direction. When two , “identical” DNA chains are laid side by side, but in opposite directions then opposing A’s and T’s can couple to each other via two hydrogen bonds, while opposing C’s and G’s can couple to each other via three hydrogen bonds. The “code” carried by either chain of bases provides our genetic information. The code from a specific stretch of DNA, called a GENE, is most commonly used in a cell to make a specific protein. All proteins are made up of a combination of twenty different AMINO ACIDS. With four letters available from the DNA it takes a set of three adjacent DNA letters to specific a single amino acid ( four squared is only 16 , while four cubed is 64 which is larger than the required value of twenty. With 64 choices some amino acids have a number of possible specifying codes. After multiple uses for a single base some combinations of letters are left over to provide STOP signals at the “end “ of the gene.) The actual “length” of a gene in base pairs is often indeterminate with sometimes 90% or more of the code between the “start” and the “end” NOT being used to actually specific the resultant protein. The “active” part of the code almost always follows a header section which may have no clearly marked - at least to human beings - starting point. Within the active region the “real” code may be broken up into a large number, say 2, or 10, or 27, or 82, different coded regions called EXONS. Each exon is separated by spacer sections known as INTRONS. Again after the “active region” there will normally be a tail section of somewhat indeterminate length. GENES A few statistics: A. 80% of human genes are closely related to “equivalent” mouse genes. B. 98.5 % of human genes are closely related to “equivalent” chimpanzee genes. The simple minded argument is that each gene has the possibility of being converted into a protein or a sub-part of a protein. In practice it takes a 50 page chapter to begin to explore some of the real-life complications in such a conversion. The first step from a segment of DNA in the nucleus to a protein consists of TRANSCRIPTION from one strand of the DNA to produce messenger RNA or mRMA. Then, after suitable cutting and splicing to remove the introns, the mRNA segment must be TRANSPORTED from the nucleus into the cytosol, then it must be TRANSLATED from mRNA into the “linear” form of the protein. Further processing FOLDS the protein into its final three dimensional form and attaches various ADDRESS LABELS to it that allow it be transported to its final point of use either within the cell or elsewhere in the body. This whole process is known as the EXPRESSION of the gene. Each step in this process may require the aid, attention, assistance, or otherwise engagement of a large number of previously produced “helpers” such as proteins, enzymes, molecules, or segments of other RNA material. For example for the transcription of a particular segment of DNA to be possible an external protein, or proteins, which in this case serves a special purpose, may first have to attach itself to the DNA, usually somewhere in the header region. In this case this external protein, or proteins, would be known as a TRANSCRIPTION FACTOR(S). (In some cases TF’s may enhance, promote, or speed-up the transcription process; while in other cases they may slow down or stop the process.) Since the same protein may serve as a TF for a number of different genes one has the possibility of one gene CONTROLLING the output from a number of other genes.(Some TF’s are believed to control as many as 2500 other genes.) If these controlled genes are expressed during the development phase of an embryo then the controlling TF may exert a large effect upon the resultant creature. MUTATIONS Any “error” in the coding, decoding, or handling processes of DNA may cause a DNA change known as a MUTATION. A mutated gene may usually be expected to produce a different protein from its un-mutated form. Most of the errors involved in the protein making process are either self-corrected by the process or don’t matter since the amount of protein produced by a given segment of mRNA is limited. Mutations caused in the DNA itself (by one of a number of possible mechanisms) may produce proteins that differ in only one amino acid or that are almost completely different from the unmutated protein. The results may be fatal, inconsequential, or somewhere in between. As nearly as one can tell mutations are the driving force behind evolution. (The business of separating the DNA and replicating it during the production of a fertilized egg is full of opportunities for insertions, depletions, crossovers, and other sources of mutations.) If the mutation is not fatal then it may result in some change or changes to some specific organ, to some nerves or muscles, or to some combinations of physical results. The argument goes that if the person with the mutated version manages to grow up, and if the changes are either “beneficial”, or at least not too detrimental, then the person carrying them may mate and pass them on to some of his or her children. (A mutation that produced a gene for red hair might be one that while not necessarily beneficial is in itself not harmful.) A single, recognizably beneficial, new gene might spread through a relatively small group within five to ten generations. Some other final effect might require repeated mutations spread out over perhaps a 1000 generations A DIGRESSION to CONSIDER the FORMATION of “MODERN” SIGN LANGUAGES A sign language is a collection of signs and expressions, often assembled in groups called “concepts”. These body movements and then used for communication among a group of sighted people. As a true language this approach allows the discussion of any topic, the telling of stories or jokes, or the transmission of poetry. Various studies, from about 1700, up to the present indicate that if a group of deaf children are allowed to get together and to remain together, even though they may have been previously isolated, and are illiterate, that they will “invent” the beginnings of a sign language. If additional young, deaf children are introduced into the group they will pick-up this initial language and will reasonably quickly refine it into a true language capable of handling any idea. Neither the vocabulary nor the grammar of this new language may bear any resemblance to the spoken language of the area in question, or to any other sign language. While there are a number of earlier records of this “language invention” happening in various parts of the world it was not until the late 1980's that it was recognized by most linguists as a possibility. The proposed dates for the “origin” of “nearly modern” man range from about 400 kya to about 120 kya. While these dates come from skulls, or partial skulls, there seems to be an almost complete lack of other fossil material. (For example material from the vertebrae that would let one see whether the side holes, that carry nerve fibers from the brain to the chest, have become enlarged from those present in chimpanzees and in our earlier ancestors. This enlargement, and thse “new” nerve pathways, allow us to control our breathing in such a manner that it allows us a wider repertory of possible sounds.) GENES, MUTATIONS, and PROTEINS In about the last four years a wealth of material has been discovered about a version of a gene on the 7th chromosome, now known as FOXP2. This is an “old gene” in that it occurs in almost the same form in all mammals from mice, aardvarks, whales, zebras, lemurs, sea lions, to monkeys, orangutans, chimpanzees, and man. It also exists in birds such as chickens. (FOX denotes a family of genes, with “branches” indicated by letters from A through Q. There seems to be some relationship between the complexity of the final “product and the number of FOX genes. Yeast has only four types, fruit flies have twenty types, and humans are believed to have at least forty types.) Human beings share a amino acid change in position 325 of the FOXP2 protein with carnivores such as cats, dogs, bears, seals, and sea lions. The exact relationship between this change and meat eating, if any, is not yet known The amino acid in position 303 of FOXP2 is the same ( THR or Threonine) in all of the aforementioned animals and birds, except for present day man where it is different ( ASN or Asparagine). FOXP2 is a controlling gene in that it serves as a transcription factor for an unknown number of other genes. It is known to have control in embryos of developments in the brain, lungs, intestines, and heart. In the usual situation each human being inherits one copy of this gene from each parent so that both strands of their DNA, in the 7th chromosome, have a working copy of FOXP2. The significance of a change in a single amino acid in a “transcription factor” protein is that it may cause a serious changes in the final “folded” shape of the protein, and therefore upon which other genes it may control Why this information is relevant to our discussion is that a three generation family in England has at least 15 of its members that have an identical mutation in one copy of their FOXP2 gene. (A change in the DNA of a G into an A in exon 14. This results in an amino acid change of an ARG or Arginine into a HIS or Histidine at position 553 of the protein.) (No other case has been found where the amino acid at position 553 of the FOXP2 protein is anything other than Arginine.) What is different about these 15 people with the “mutated” protein is that they all have similar serious speech articulation and understanding problems as well as serious problems with grammar. fMRI studies of the brains of these “affected” people produce markedly different patterns of activity, from “unaffected” people. These show up particularly when engaged in either silent or spoken language tasks. These operations included repeating a word or indicating a verb to go with a given noun. These studies indicate that the areas in the left brain that normally deal with speech and grammar problems, even in users of sign languages or Braille, are not properly developed in people with a mutated FOXP2 gene. Apparently a single working copy of the FOXP2 gene is sufficient for the correct development of the lungs, heart, intestines, and other lower body parts that may be controlled by it, however a single un-mutated version is not sufficient for the normal development of some important parts of the brain. It turns out that by various techniques the people working on FOXP2 have been able to place time limits upon the mutation that separates present day man’s version from the chimpanzee’s version. It is presently believed that the significant mutation occurred less than 200 kya and more than 50 kya. A POSSIBLE SCENARIO FOR THE ORIGIN OF HUMAN SPEECH. Since all other primates can make sounds and use cries and gestures to communicate with each other it is rational to assume that our ancestors such as Homo erectus or Homo ergaster or Homo heidelbergensis did also. As their physical form changed their ability to make various sounds may have increased which may in turn have increased their vocabularies over those of their tree dwelling relatives Just as chimpanzee and gorilla infants still learn the “languages” and “accents” of their particular group one assumes that “human-like” infants also learned theirs the same way over the some millions of years that proceeded Homo sapiens. If one assumes that certain bodily changes such as the placement of the larynx and the expansion of the nerve network controlling the chest had already taken place by 70 kya, then the key mutation of the FOXP2, in a single person, might have served as the missing link for the production of human speech. The path involves a number of assumptions, however so does every other possibility. A. The mutation of FOXP2 into the “human” form had to occur in a single person. B. The “carrier” of the new gene had to survive and reproduce. C. At least some of his or her children had to survive and reproduce. D. At least one set of children or grandchildren, both of whom had inherited the new gene had to survive to maturity and then mate and reproduce. E. Eventually some offspring had to inherit two copies of the new FOXP2. Father-daughter or son-mother mating would offer a quicker route to “two copy” offspring. Both types of such matings have been observed among various groups of present day chimpanzees. F. Some of these “two copy” offspring had to mate and reproduce. G. Etc.,etc., for several more generations. H. Once “two copy” offspring became common the probable tendency of most of them to mate with other “two copy” individuals would have eventually led to the disappearance of the non- two carrier population. The outline above would be most likely if the group involved was relatively small and if they did not have any strong taboos against incest It would probably be speeded-up by positive feedback as the first and second generation “carriers” exhibited superior vocal skills and became likely to seek mates with similar skills. The positive feedback possibility assumes that “half a loaf is better then none” in that the impaired speech possibilities of the carriers of a single copy of the new gene would be quite superior to those members of the family with no copies. When more single or double copy babies were born we assume that these infants would learn their mother or father tongue just as present day children or chimpanzees do. Based upon our present day knowledge of sign language “invention” one would expect that when “two carrier” babies were eventually born they would learn the existing set of noises and would then improve upon them. While the first one carrier people might have been restricted to sounds for nouns such as water, fire, lion or antelope, their children, even if still only one carrier, would probably have extended the vocabulary to more complicated combinations. By the time one reached the stage of some number of two mutation carriers one would further expect that single word expressions had been combined into phases such as “ you get water”, “ come this way”, or “help me catch this pig” By the time the next generation appears it seems possible that with a larger vocabulary and inherently better language skills some youngster would have strung together a number of “words” to get an expression such as “Wouldn’t be nice to use this red pigment to paint our faces.” Once an expression of this complexity had been accepted and generalized then the group would have been off and running toward a full-fledged language. With generations repeating on a 15-20 year cycle the complete change from “no language” to “working language” could have occurred in an evolutionary “instant” such as several hundred years. “AFTER THE FACT” INSIGHTS CONCERNING THE STEP FROM SPOKEN SPEECH TO WRITING While spoken language may have originated about 50 kya we have only the vaguest ideas about either its pronunciation or its structure. There are societies with a strong oral traditions that may sometimes have passed some stories and songs down for several thousand years or more, perhaps almost unchanged in content or sound, however 45 thousand years of oral tradition seems unlikely. Thus when we speak of history we normally means recorded history, which in turn implies the existence of a writing system. While writing is now reasonably widespread as far as we now know only four prehistoric societies ever crossed this intellectual boundary by learning how to record speech in enough detail to be able to describe almost any event or idea. These four were the Sumerians and the Egyptians in the Middle East, the Chinese in Asia, and the Maya in Middle America. The physical symbols invented for the writing system in these four cases were all quite different. All but the Chinese system have now been out of use for hundreds or thousands of years. When they were rediscovered, during the last two hundred years, they posed difficult decipherment problems that have not yet been completely solved. As far as we know almost all existing written languages are descendants of writing systems started by the Egyptians, Sumerians or the Chinese. In the beginning the number of readers was probably about equal to the number of writers. In terms of commercial transactions or taxation matters the reader and the writer were often the same person, the local scribe. In early religious writings it may have been assumed that the gods could read the messages directed to them and no human readers were necessary or expected. The symbols invented by the four societies that, apparently independently invented writing, are the subjects of some of the following chapters. As we shall see the initial transition from the “no writing” state to the “writing” state was not a step function. While our information is sketchy this transition normally took several thousand years. At its start this process had no clear objective, nor was it the result of planning by a single person, group, or committee. In all the cases that we shall consider our knowledge is limited partially by the fact that we only get glimpses of the progress at irregular intervals in time and at particular places. Neither these times nor these places may be those where the most significant advances were being made. With the benefit of hindsight it now appears that for writing to develop in a society, as a minimum, it needed to have: A. Some fixed settlements. B. A reasonably complex social political organization. C. Facilities for the production of painted pottery D. Enough technology to support the writing system that was set up. E. Some influential segment of the population that had,or perceived that it had, some void that this new gimmick would fill. Many societies have existed that had the first four properties but that never developed a system of writing. That is, while these four conditions appear to have been necessary, without the last condition they were not sufficient. What is noticeably different among the four cases being considered is the apparent “need” that was filled by the use of writing. In the Sumerian case it was the recording of lists and numbers to allow what we would call accounting practices. In the Mayan case it was also for the recording of numbers, however in this case it was for recording astronomical information or the dates of significant occurrences. In the Egyptian case it seems to have been to record ownership of goods with names, while in the Chinese case the earliest impetus, or which we are aware, seems to have been to record the results of forecasts made by the interpretation of cracks in the so called oracle bones. As far as we can tell all of the societies that nurtured full blown writing systems had counting systems in place before their writing became at all generalized. Since all the numbering systems were different they all had different spoken speech approaches to a given number. In several cases we have artifacts showing dates or other uses of numbers hundreds of years before more generalized writing appears. While these examples are certainly precursors of writing they are not writing in the sense that we are considering it and their use in a society does not necessarily forecast that this society will end up producing a writing system, since it often did not happen. POSSIBLE METHODS OF CONVERTING SPOKEN SPEECH INTO WRITTEN SPEECH Spoken speech, of a certain language, consists of sound patterns that have had their meanings agreed upon by some group of people, known as the speakers of that language. If writing is to record a segment of this speech and allow it to later be converted back into the spoken language then there must be agreement on a set of written symbols that will record specific sounds. While there may be a number of theoretically possible ways to make up symbols to use in writing, only four have ever been widely used. One can deal in “words”, one can deal in “syllables”, one can deal in “morphemes”, or one can deal with letters. (Some American Indian languages have some one sentence words, however we shall ignore them for the moment.) In practice writing had been around in Sumer and Egypt for at least 1500 years before we have any knowledge of an alphabetic script such as Ugaritic. (Phoenician came from Ugaritic and then about 600 BCE written Greek came from Phoenician. Aramaic, Hebrew, and Arabic, among others also came from Phoenician.) Ugaritic was written using cuneiform, Phoenician used the Ugaritic alphabetic concept and changed the symbols. If the spoken speech has separate, “one idea” words then one could have one symbol for every word. If the speech has plurals such as “cow” and “cows” and verb constructions such as “talk”, “talks”, “talked”, and “talking” that each require a separate, and perhaps unrelated, symbol one can see that the number of required symbols may be very large indeed. Such an approach would be called logographic with word based symbols. While all the early languages may have started off in this direction they reasonably quickly modified their approach so that at least some symbols stood for syllables, syl-la-bles, or morphemes instead of words. While one symbol per word might lead to 50,000 symbols, in most languages 500 to 600 symbols for syllables should be sufficient to deal with 50,000 words. There is in fact a special form of old Persian that was developed for King Darius that used only 36 syllabic symbols to deal with the whole language. In most languages, there is a separate set of signs for numbers. In practice many words may have only one syllable (Some English examples include: and, but, cat, dog, fox, girl, have, inch, jest, knock, law, make, name, old, purse, quick, red, stop, tea, used, vain, watch, xyst, yacht, and zinc.) Thus with a syllabic writing system there will still be some symbols that can stand for complete words. A third approach, of which Chinese turns out to be a prime example, is to have one character per morpheme. (From a linguistic viewpoint a morpheme is defined as the smallest meaningful unit in the grammar of a given language. In English a given word might have three morphemes and four syllables, or two morphemes and one syllable, or one morpheme and two syllables. For example in English cats has only one syllable, but two morphemes, “cat” and “s” while unladylike has three morphemes and four syllables. As illustrated by the morpheme “s” an English morpheme does not have to be a whole syllable. In Chinese essentially all morphemes are single syllables. Just as in English a single syllable could make up a whole word, so in Chinese a single morpheme represented by a single character can represent a whole “word” (Since Chinese is made up a string of characters that present some possible train of thought the concept of “word” is not necessarily the same as it is in English.) In practice most Chinese “words” require at least two symbols. Except for personal pronouns Chinese does not have plurals, in the Western sense. Thus one may say “many cow”, “few cow”, or “ten cow” . In general Chinese verbs have no “tense”, neither do English verbs such as “cut” or “put”. While some reasonably early languages provided some form of word “dividers” many did not. Written English only got around to them around 1000 CE and they are a rarity in present day written Chinese. |
Seeking Agent |
||
|
