15. Example cases of CMM applications

The use of clay for CMM in ancient Mesopotamia is an instructive case. The material situation was characterized by a shortage of the essential construction materials wood and stone. Most construction was done with sun dried clay bricks, and composite construction in daub and whattle, i.e. a reed and clay composite material. Clay was used for all imaginable objects of daily use, even furniture and doors. Wooden doors were rare prestigious objects which were handed down as heirloom and dowry, bequeathed in wills, and were regarded as family treasures (Vajda 1995: 23). Since one needs wood to fire clay (but that was not available), fired and glazed tiles were reserved for the most prestigious state buildings, like the Ishtar gate in Bablyon. (Personal communication: Dr. Alfred Schinz .)

As was noted above, the economic advantage of clay is that it is abundant and cheap to procure, easy to inscribe, and best of all, very durable when baked, but because of the shortage of firewood, most of the clay for recordkeeping was just sun-dried. The information- density-/ stability- / durability / -weight tradeoffs of clay are this : It can only be handled in small bricks (because larger ones would break and/or be too heavy to handle) . The maximum size of a clay tablet that can be handled by a person is about 30*30 cm [480]. Information density is similar to Egyptian writing with about 3 chars per cm ² , giving roughly 1000 chars on a 20*20 slate. A 10*10 slate, which is about the optimum size for handling in wet form, contains only about 300 chars. So while the information density per unit of storage may not seem to compare too badly to the Egyptian information technology, the information/weight ratio is dismal: about 1/5 to 1 kg per 1000 chars, compared to about 10 to 50 gr. for the papyrus. The material stability and handling properties are, overall, not too advantageous, either: to transport the equivalent of a book (1000 slates = 200 to 1000 kg), one needed a sizeable oxcart or two of them. And one better didn't move the clay tablets at all, for danger of breaking them or getting them in disorder. The mechanical clumsiness of clay bricks severely restricted the collation of larger volumes of clay inscribed tablets to books. One clay "book" needed a whole storage rack, built quite solidly (of the very scarce wood), which would otherwise hold a whole library of papyrus rolls. Compared with an Egyptian papyrus scroll library, [481] in clay tablets, would need several large rooms instead of just one. But, as already mentioned, clay has advantages that make it specifically useful for purposes of which the erstwhile babylonian writers and administrators would have been hard put to think of: The fact that the Ashurbanipal library was burned luckily preserved the originally unbaked clay tablets for eternity. This effect was quite contrary to what happened to the Alexandria library. By this quirk of antique information technology, was it possible at all for modern archeology, to find out so much about the ancient cultures of Mesopotamia and other places of the Near East, who also used clay tablets, like Ugarit and Mycenean Greece, and which were also gutted by fire.

Besides clay, the Mesopotamians used waxed wood for short and medium time record keeping. Since Assyrian times, papyrus was also used, since Assur had invaded and conquered Egypt, thus opening up the papyrus trade. Because of the particularities of the writing technique, the cuneiform writing system can only be used with soft malleable surfaces. It is unsuitable for papyrus and ink writing. For this reason, the scribes in Assyruan and Persian services also used the Aramaic language and Aramaic characters (Kerckhove 1988).

Papyrus as material for CMM is a historically unique case, as it is native to the Nile river (only later was it transferred to other regions), and the combination of the Nile swamps as the habitat of the plant, and the very dry surrounding climate of Egypt provided a singular situation that allowed the development and use of that otherwise very perishable vegetable fiber substance for writing material. Papyrus would deteriorate quickly in moist climates, like for example India. Moreover, the ecological situation of ancient Egypt gives another singularity, since a dry climate usually cannot support a high population density (and support a level of cultural organization that gives the incentive to develop a writing system) such as was possible in Egypt without much effort through the Nile floodings. The other historical case of early civilization, Mesopotamia, achieved that end only through the application of mass labor for vast irrigation systems.

Papyrus was the standard record keeping material in ancient Egypt, and later hellenistic Greece and Rome. But even in a dry climate, papyrus ages and needs to be replaced, the contents of a roll needs to be copied onto a new roll. The information-technical infrastructure and requirements for the upkeep of a papyrus library is sketched under the header: "A model papyrus scroll library", p. 170. Its later use in the different climates of Greece and the Roman empire suffered from its low durability in humid conditions. Later, the much more durable parchment was substituted, but parchment is so much more expensive, giving rise to other information problems, as discussed by Innis (1972: 85-140).

A library of about 10,000 scrolls could be called representative for a large regional administrative center or temple in the late Egyptian / Hellenistic world. About 20 scribes would be the personnel necessary for the upkeep and maintenance of the library. As theoretical maximum, one scribe can copy at most two scrolls per day, so that the whole library could be recycled in about 15 years by one person. But the practical situation would be much less efficient. The repeated copying of papyri allowed the transmission of the records through three millennia.

The end of the dominant role of the Papyrus CMM is marked by the destruction of the Alexandria Library, which also marked the fall of the whole of antique culture. Canfora (1988). Fire was the deadly enemy of this CMM (same as paper), and before the adaptation of fire-resistant buildings, fire extinguishing technology and the fire brigade, the paper/payrus based CMM were extremely vulnerable to physical destruction, leading to an extreme fragility of writing based civilization. Platon's criticism of writing in Phaidros (274c) [482] had indirectly hinted at some of these problem factors: the relegation to writing makes the CMM prone to accidental loss and vulnerable to accidental destruction. Oral memory is protected by the self-preservation instinct of its bearer. Also, most oral material is stored redundantly, because in the oral traditions, everything that is memorized is recited aloud, usually in front of an audience, and many times. The natural tendency of humans to tell, and listen to interesting stories, is the life blood (or the driving machinery) of oral tradition. Ong (1977: 30-77). This insured that many people of a society had heard at least some part of the story, and could retrieve it out of their own memory storage, even if the Aoide died without a successor. The collating of many distributed pieces of collective memory is probably how the "Homeric" epics as we know them came into written existence. Illich (1988: 15-21), (Powell 1991), Ong (1982: 58-61). Only when book printing made the physical destruction of all existing exemplars of a text less likely, could writing based civilization really take hold. Goody (1977), (1986), (1987), Ong (1977), (1982), Scharlau (1986).

In Antiquity, the connection of papyrus to the large scale bureaucratic record keeping requirements of the ancient empires was a decisive factor, as Innis has noted (1972). The Roman Empire needed a constant supply of papyrus for its bureaucracy, and its lack contributed to the collapse. Mcluhan (1972: 61). [483] Indigenous cultures did not only resist the adaptation of writing out of sheer primitivity, but also because writing enforces a certain cost factor, its insidious main problem being that what is stored on paper (or other materials) will tend to vanish from human memory, as Platon had duly noted in Phaidros. The huge knowledge loss after the breakdown of the ancient civilizations was inevitable because of lack of skilled manpower to maintain the libraries, since papyrus has a half-life time of about 100 years in a humid climate like the northern Mediterranean areas, and would simply rot away, even if the libraries had not been burned. To keep a library operative, a constant expenditure in man-power was needed for the re-copying of scrolls. In ancient times, this work was mostly done by slaves. This highly trained slave manpower was not available any more after the breakdown of the Roman empire, so nothing could be done to prevent the massive information loss. Finally, the medieval church reconstructed the manpower -training and -copying infrastructure based on the parchment and on the monastery system to revive writing culture. The only place where papyrus naturally survives for long periods, without recopying, is in fact Egypt, because of the extreme dryness of the country.

This is a calculation of the information infrastructure requirements for papyrus libraries. For lack of original sources, the comparable figures of a medieval monastery library are taken as base. Illich (1988: 37-42). The largest known ancient library was that of Alexandria in Hellenistic times, it had 500,000 or more scrolls (Canfora 1988). An average regional center in Hellenistic or Roman times may hold a library consisting of about 10,000 scrolls [484]. The papyrus scrolls came in different sizes: from 10 cm to 30 cm in width, and 3 m to 9 m in length. The optimum size is determined by material properties and the conditions of writing: scribes squatted down on the ground, and the skirt of the scribe served as writing table substitute. An extensive discussion of the psychological and physiological factors of writing is in Kerckhove (1988): "The Alphabet and the Brain". Main references there: Sirat (1988: 173-201), Skoyles (1988: 362-380). A scroll 9 m in length is about the maximum that can be handled with relative ease, although much longer ceremonial or burial scrolls are known that were not intended for daily use. A long scroll is impractical for text that must be referenced often, because of material wear.

To facilitate the informational calculation, we will commit a little anachronism and apply the DIN norm to ancient Egypt. On the average, a scroll would be 3 m in length and 30 cm in width, which is about the equivalent of 15 pages DIN A4 paper per scroll. Let us calculate 10 min. scribe time per A4 page, at a yield of 1000 chars per page. This gives us 15,000 chars per scroll, needing 150 min. or roughly 3 hours (with a break) to copy a scroll. A modern book with 300 pages and the higher printing density of about 3000 chars per page contains about 1,000,000 chars, ie. the equivalent of 60 papyrus scrolls. If we have 10,000 scrolls in our egyptian model library, this would contain in total 15,000 * 10,000 = 150 million (mega) chars, or 150 books. This is about what an ordinary literate citizen in our contemporary society may have sitting on her private bookshelf (requiring 3 cm shelfspace per book, totalling 450 cm, i.e. a shelf with 5 boards of 1 m length), and even if she doesn't keep them around her apartment, this is a good measure of what amount of reading material someone who does very little reading, will process in a whole life - if it is not books, then it will be newspapers and magazines. For a modern literate worker, 150 books or 150 mega chars may be a measure of what is routinely processed (read or scanned and excerpted) in about one year, for a research project, or a dissertation.

To return to our ancient model library: To write one scroll, we need about 3 hours. To write 10,000, we need 30,000 hours. There is considerable overhead for handling the scrolls, and doing supportive tasks for the writing process. Because of the low information density, scroll handling is expensive. One scroll, when rolled up, is a cylinder of about 5 cm diameter and 30 cm length. A rack to orderly store single scrolls would need a "pigeonhole" construction of about 10 cm squares when done in wood. For human access, we limit rack height to 1.5 m or 15 stack spaces in height. That amounts to 667 stack spaces in width, or about 66 m, to store 10,000 of them. Compared to our book library, that is a large size storage for the amount of information stored. To house it, we need a room of about 20 m length and 6 m width, with two double racks. Since one would more likely keep the scrolls in batches of 10 to 20 in a 20*20 cm cubicle, the rack length required would be less, about 15 to 25 m. Still there is some footwork to do to retrieve the scrolls from the racks. Scrolls need to be flattened out to read. The flattening procedure adds another cost factor into egyptian information processing: This becomes clear when compared to a book. The book binding allows a relatively sturdy back material to hold together the soft and thin pages. When reading one page, one doesn't need to disturb the pages lying in front or back. In order to read through a scroll, there will be mechanical wear and tear on the text portions that have to be unrolled before reading the one that is looked for. This adds another factor shortening the life of a scroll. König (1997,I: 158-159, 312-313).

Writing material needs to be prepared, there is no factory made pen and ink available, which severely limits writing speed when you have to prepare your own ink and chew a reed stick that you call your 'pen' every few minutes. Let us calculate six hours per roll including support work and breaks, that gives 2 scrolls per working day of 12 hours. Let us assume 300 working days per year per scribe. To copy the whole library of 10,000 rolls, he needs about 15 years. Because after 15 years, the rolls which he had copied at the start of his career, would have worn out and need to be replaced by now, so he could start afresh copying his library. Now actual turnover and wear and tear of papyrus is probably higher in constant use, so that we should multiply the man- (scribe-) power requirements of ancient Egyptian word processing by a factor of ten. Ten scribes to support just the permanent data base. How many more for day-to-day and record keeping tasks that the organization requires? The libraries resided in the temples. These were also the center of all the accounting and bookkeping of the whole region, starting with the regular work of keeping track of fields between the Nile floods, the inflow and outflow of seed and foodstuffs in granaries and storehouses to support the local economy, and the tributes paid to the higher centers: the central temples, the pharaonic tributes. Then add the long term requirements for building and transport and irrigation infrastructure construction and maintenance, waterways, channels, flood gates, etc. The number of scribes required is hard to calculate, but we have figures that are from much more recent times, the european monastery system. It needs to be remembered that up to Gutenberg's time, 1450, the information processing situation all over the world was not very different from the Egyptian scenario. A given region may be able to support a monastery housing 100 monks, and when there are 20 to 30 specialist scribes among them, this will be a fairly accurate figure. Illich (1988: 37), (p. 39-40) for the problems of indexing and retrieval of scripts.

In a hot humid climate like India, the durability problems of materials like papyrus are considerable, especially in the monsoon time. (Sandermann 1997: 232-234, and personal experience ). The monsoon wet rot and insect appetites would have considerably shortened the half-life of papyrus scrolls und made it difficult to impossible to keep large papyrus libraries in airy wooden buildings open to the insects. Here, the first papyri would have rotten away before the scribes had been through copying the last one [485]. It is likely that this was a very practical reason why the Indian Brahmin culture preserved its oral memory aspect so much more dilligently than the ancient near-eastern cultures in their drier climates.

In hot and humid climates, the long-time storage of large-volume papyrus or paper libraries needs buildings of stone or brick construction. This is again a matter of social investment for the necessary buildings. Stone construction is more expensive than wood, and brick has become more widely available in modern times only. To bake bricks, even though the clay source material may be abundant, needs a lot of thermal energy. From the wood needed to fire the bricks for one building, many houses could be constructed. The availability of wood (or the scarcity thereof) was a serious limiting factor. In rural India, the prime heating material for cooking was/is cow dung, which doesn't yield enough energy density for brick firing. Harris (1989: 309-313). Before the advent of coal mining, the energy cost of bricks was therefore prohibitively high . Even in mud-rich mesopotamia, most of the construction was done with sun-dried bricks, and only a few glazed tiles like they have been preserved in the outer covering of monumental builidings .

The studies on indigenous use of color in CMM, Silverman (1991), Barthel (1971), of quipus and Andean weaving patterns, quillcas, indicate that the informational dimension of color was utilized in more depth and detail than in the European CMT. European {writing / printing} is largely color insensitive (black / white) because of technology. Since color printing is more expensive than black and white, this informational dimension tends to be used less in the western CMT after the printing revolution. Illich (1988: 50) notes that color was widely used in earlier European manuscripts [486] and even more so in Arabic writing. A notable case of extenxive color use in Western CMT is of course art, and the craft of map making. Tufte (1990), (1992) gives ample information on the informational usage of color, especially where it can convey information "at a glance" extremely quickly, through judicious usage of the eye's vast capacity of color processing, that is entirely cut out in the black/white alphabetic technology. With modern multimedia computer technology, color begins to be used on a wider scale, but there is little knowledge how to apply it wisely. Tufte (1990: 82, 88) describes some of the garish misuse of color by aesthetically untrained software designers who have suddenly the color palettes of computer video available, but don't quite know how to use them.

When the European and South American cultures clashed in the conquista, there arose many issues of cultural domination. Besides the brute dominance factor of "guns, germs and steel" (Diamond 1997: 67-82), an important question is whether the Inca quipu system could be shown to have qualities that the European writing system lacked. This (so far hypothetical) factor is here called the quipucamayoc mnemotechnics. The further question to ask is, if the prior discussions in the anthropological literature of the quipu may have systematically omitted relevant material .[487] The prior interpretations focus on the aspect that quipu may be limited in its semantic expressibility because it cannot encode language. Scharlau (1986: 89): "... Grenzen. All das was nicht in irgendeiner Weise quantifizierbar ist, kann offenbar nicht dadurch festgehalten werden." This may contain a misconception that could have been created as consequence of the specific attention that the Spaniards (as the not-so-objective-and-impartial documentors of the last remaining traces of quipu mnemotechnics) concentrated on the numeric statistical value of the quipu as devices to extract tribute and taxes from the population. It is easy to understand that the Spaniards, after destroying all the quipus they could locate (and probably killing all the quipucamayoc they found as well), (Scharlau 1986: 84), then belatedly noticed that after they had carried off all the booty and loot from their conquista, that the administration of their newly acquired territories suffered terribly from lack of statistical information, and then they were told by the natives that this had worked perfectly well with the quipu system that the conquistadores had just destroyed so diligently. So the Spaniards tried to preserve the last bits of the knowledge that was encoded in them for the purpose of extorting the tributes like the Inca rulers had done so expertly before them. Engl (1991: 156, 203-204, 359-370, 400-406), Scharlau (1986: 89-90).

We will now ask the converse question: what could have been the special advantage of the quipu mnemotechnics , which the quipucamayoc used, and could this have been something beyond the expressive range of alphabetic writing? Scharlau makes reference to Guaman Poma de Ayala, who attempts to describe in his work the old quipu-order with the new medium of alphabetic writing (1986: 92). Of course , Guaman as a member of the conquered society was obviously not able to stand up and declare that the categorical ordering system developed by the quipucamayoc that was embodied in their mental operation modes , (or their CMA in the present context), was superior to anything that the Spaniards could use, and was unparalleled in all symbolic productions of humanity, until the 20th century, when structural formal grammars were invented and employed in computer software technology. The strands of the quipu could be used for categorical ordering systems that are described with technical formal grammars. See Bauer (1971,II: 100-144), Brauer (1968: 108-115). It can only be inferred indirectly that the quipu may have been the ideal medium for a structural categorical notation and (perhaps) it could have been developed further in this direction towards a superior structural categorical mentality , had the Spaniards not cut the whole culture short . Scharlau (1986: 84).

Another possibly decisive mnemonic aspect of the quipu that has not been found in the alphabetic literature, is the tactile aspect of quipu production and reading . It is mentioned in the respective section that western cultures tend to repress the tactile dimension. The example case that demonstrates this glaringly, may be that the most conspicuous use of knotted cords in the Eurasian civilizations was for flagellation. [488] This aspect of the tactile dimension, because it has not been found in any literature, cannot be discussed directly, but some indirect reasoning may be allowed. The tactile dimension of quipu reading gave it a great difference over the visual alphabetic technique, that has been described as an excarnated CMT. [489] The possible effect of a much more efficient enhancement of the personal memory recollection by the multi-medial, poly-aisthetic effect of something that is felt simultaneously when it is seen as form and as color, and recited aloud, while memorizing it, can only be hinted at with the observation of Schärli and the work of the Polyaisthesis school. Schärli (1996: 29), Roscher (1997) and Platon's famous passage in Phaidros. The important mnemonic effect of the touching is again evidenced by the universal usage of the rosary.

[480] The size of modern roof tiles indicates the optimal size/weight/stability distribution that can be achieved with high quality oven fired clay. Ancient Mesopotamian technology (where clay was mainly used for writing) was in no position to supply such a consistent material engineering standard, mainly because of energy scarcity. One had to burn whatever organic matter was available, like cow dung in contemporary India. This doesn't yield the high temperatures to harden the clay completely.

15. Example cases of CMM applications

15.1. The use of clay for CMM in ancient Mesopotamia

15.2. Papyrus use as CMM in Antiquity

15.3. A model papyrus scroll library

15.4. Cultural memory, civilization, and climate

15.5. Color as coding dimension

15.6. Quipu mnemotechnics

15.6.1. The tactile aspect of quipu production and reading