Chapter XII: Part III: Graphic Systems and Literature (1)

INTRODUCTORY.

The intellectual development of a nation attains its fullest expression in language, oral or written. This “divine art” as Plato calls it, claims therefore from the student of man in the aggregate a prolonged attention and the most painstaking analysis. Too frequently one hears among anthropologists the claims of linguistics decried, and the many blunders and over-hasty generalizations of philologists quoted as good reasons for the neglect or distrust of their branch.

The real reason of this attitude I believe to be not so much the mistakes of the linguists, as a strong aversion which I have noticed in many distinguished teachers of physical science to the study of language and the philosophy of expression. The subject is difficult and distasteful to them. Having no aptitude for it, nor real acquaintance with it, they condemn it as of small value and of doubtful results. I have never known a scientific man who was really a well-read philologist who thus under-estimated the position of linguistics in the scheme of anthropology; but I have known many who, not having such thorough knowledge, depreciated its value in others.

The third and fourth parts of this volume are devoted to language, the third as it appears especially in its written forms, the fourth particularly to the profounder questions of linguistic philosophy. Here again I shall be found in opposition to the majority who have written on these subjects. The claim I make for the largely phonetic character of the Mexican and Maya hieroglyphs is not generally accepted; and the poetical spirit which I argue exists in many productions of the aboriginal muse will not be favored by those who deny the higher sentiments of humanity to uncivilized man.

I have endeavored by frequent illustration, and reference to the best sources of information, to put the reader in the position to judge for himself; and I shall feel highly gratified if he is prompted to such investigations by what I may say, whether his final conclusions agree with mine or not.

THE PHONETIC ELEMENTS IN THE GRAPHIC SYSTEMS OF THE MAYAS AND
MEXICANS.[201]

All who have read the wonderful story of the Spanish conquest of Mexico and Central America will remember that the European invaders came upon various nations who were well acquainted with some method of writing, who were skilled in the manufacture of parchment and paper, and who filled thousands of volumes formed of these materials with the records of their history, the theories of their sciences, and the traditions of their theologies. Aiming at greater permanence than these perishable materials would offer, they also inscribed on plinths of stone, on slabs of hard wood, and on terra cotta tablets, the designs and figures which in the system they adopted served to convey the ideas they wished to transmit to posterity.

In spite of the deliberate and wholesale destruction of these records at the conquest, and their complete neglect for centuries afterwards, there still remain enough, were they collected, to form a respectably large _Corpus Inscriptionum Americanarum_. Within the present century many Mexican and Maya MSS. have for the first time been published, and the inscriptions on the temples of southern Mexico and Yucatan have been brought to the tables of students by photography and casts, methods which permit no doubt as to their faithfulness.

Nor have there been lacking diligent students who have availed themselves of these facilities to search for the lost key to these mysterious records. It is a pleasure to mention the names of Thomas and Holden in the United States, of De Rosny, Aubin and de Charencey in France, of Förstemann, Seler and Schellhas in Germany, of Ramirez and Orozco in Mexico. But it must frankly be confessed that the results obtained have been inadequate and unsatisfactory. We have not yet passed the threshold of investigation.

The question which forces itself upon our attention as demanding a reply at the very outset, is whether the Aztec and Maya systems of writing were or were not, in whole or in part, _phonetic_ systems? Did they appeal, in the first instance, to the _meaning_ of the word, or to the _sound_ of the word? If to the latter—if, in other words, they were phonetic, or even partially phonetic—then it is vain to attempt any interpretation of these records without a preliminary study of the languages of the nations who were the writers. These languages must moreover be studied in the form in which they were spoken at the period of the conquest, and the course of native thought as expressed in the primitive grammatical structure must be understood and taken into account. I hasten to add that we have abundant materials for such studies.

This essential preliminary question, as to the extent of the phonetic element in the Mexican and Maya systems of writing, is that which I propose to put at present, and to answer it, so far as may be. Hitherto, the greatest diversity of opinion about it has prevailed. Some able writers, such as Valentini and Holden, have questioned the existence of any phonetic elements; but most have been willing to concede that there are such present, though their quantity and quality are by no means clearly defined.

We may assume that both systems under consideration are partly ideographic. Every system of phonetic writing introduces ideograms to some extent, our own among the number. The question is, to what extent?

But before we are prepared to answer this question about the extent of the phonetic element, we must seek to ascertain its character. We are all aware that a phonetic symbol may express the sound either of a whole word of several syllables, or of a single syllable, or of a simple acoustic element. Again, a single phonetic symbol may express several quite diverse sounds, as is familiarly exemplified in the first letter of the English alphabet, which represents three very different sounds; and, on the other hand, we may find three, four or more symbols, no wise alike in form or origin, bearing one and the same phonetic value, a fact especially familiar to Egyptologists.

We must further bear in mind that the arrangement to the eye of phonetic symbols is altogether arbitrary. Because a prefix is pronounced first in the order of time and a suffix last, it by no means follows that the order in space of their corresponding symbols shall bear any analogous relation. The idea awakened by the sound of the word is a whole, and one; and so that this sound is represented, the disposition of its component parts is, philosophically speaking, indifferent. When it is remembered that in most American languages, and notably in the Mexican or Nahuatl, there is a tendency to consolidate each phrase into a single word, the importance of this consideration is greatly increased.

As the position of the phonetic parts of the phrase-word may thus be disregarded, yet more indifferent is the order of sequence of the symbols. There is no _a priori_ reason why this should be from left to right as in English, or from right to left as in Hebrew; alternately, as in the Boustrophedon of the Greek; or from top to bottom, as in Chinese.

In such an examination as the present one, we must rid our minds of the expectation of finding the phonetic elements in some familiar form, and simply ask whether they are to be found in any form.

We are not without a trustworthy guide in this quest. It is agreed among those who have most carefully studied the subject that there is but one path by which the human mind could have originally proceeded from picture-writing or thought-writing to phonetic or sound-writing. This was through the existence of homophones and homoiophones in a language, of words with the same or similar sounds, but with diverse significations. The deliberate analysis of a language back to its phonetic elements, and the construction upon those of a series of symbols, as was accomplished for the Cherokee by the half-breed Sequoyah, has ever been the product of culture, not a process of primitive evolution.

In this primitive process the sounds which were most frequently repeated, or were otherwise most prominent to the ear, would be those first represented by a figure; and the same figure would come to be employed as an equivalent for this sound and others closely akin to it, even when they had other connections and bore other significations. Hence affixes, suffixes, and monosyllabic words, are those to which we must look as offering the earliest evidences of a connection of figure with sound.

According to the theory here very briefly indicated, I shall examine the Maya and Nahuatl systems of writing, to ascertain if they present any phonetic elements, and of what nature these are.

Turning first to the Maya, I may in passing refer to the disappointment which resulted from the publication of Landa’s alphabet by the Abbé Brasseur in 1864. Here was what seemed a complete phonetic alphabet, which should at once unlock the mysteries of the inscriptions on the temples of Yucatan and Chiapas, and enable us to interpret the script of the Dresden and other Codices. Experience proved the utter fallacy of any such hope. His work is no key to the Maya script; but it does indicate that the Maya scribes were able to assign a character to a sound, even a sound so meaningless as that of a single letter.

The failure of the Landa alphabet left many scholars total skeptics as to the phonetic values of any of the Maya characters. To name a conspicuous and recent example, Prof. Leon de Rosny, in his edition of the Codex Cortesianus, published in 1883, appends a Vocabulary of the hieratic signs as far as known; but does not include among them any phonetic signs other than Landa’s.

But if we turn to the most recent and closest students of these records, we find among them a consensus of opinion that a certain degree, though a small degree, of phoneticism must be accepted. Thus our own able representative in this branch, Prof. Cyrus Thomas, announced in 1882, in his _Study of the MS. Troano_,[202] that several of the day and month characters are, beyond doubt, occasionally phonetic.

Prof. Förstemann, of Dresden, whose work on the Dresden Codex has appeared quite recently, announces his conclusion that the Maya script is essentially ideographic;[203] but immediately adds that the numerous small figures attached to the main sign are to be considered phonetic, and no matter in what local relation they may stand to this sign, they are to be regarded either as prefixes or suffixes of the word. He does not attempt to work out their possible meaning, but, as he says, leaves that to the future.

Almost identical is the conclusion of Dr. Schellhas, whose essay on the Dresden Codex[204] is a most meritorious study. His final decision is in these words: “The Maya writing is ideographic in principle, and probably avails itself, in order to complete its ideographic hieroglyphs, of a number of fixed phonetic signs.”

FIG. 1.—The Maya Hieroglyph of the Firmament.
]

Some of these signs have been so carefully scrutinized that their phonetic value may be considered to have been determined with reasonable certainty. An interesting example is shown in Fig. 1, for the analysis of which we are indebted to Dr. Schellhas. The quadrilateral figure at the top represents the firmament. One of the squares into which it is divided portrays the sky in the day time, the other, the starry sky at night. Beneath each are white and black objects, signifying the clouds, from which falling rain is indicated by long zigzag lines. Between the clouds on the left of the figure is the well-known ideogram of the sun, on the right that of the moon. In the Maya language the sun is called _kin_, the moon _u_, and these figures are found elsewhere, not indicating these celestial bodies, but merely the phonetic values, the one of the syllable _kin_, the other of the letter _u_. The two signs given in Landa’s alphabet for the letter _u_ are really one, separated in transcription, and a variant of the figure for the moon with the wavy line beneath it. The word _u_ in Maya is the possessive adjective of the third person, and as such is employed in conjugating verbs, the Maya verbal being really a possessive.

A very common terminal syllable in Maya is _il_. It is called by grammarians “the determinative ending,” and is employed to indicate the genitive and ablative relations. Dr. Schellhas considers that this is represented by the signs affixed to the main hieroglyphs shown on Fig. 2.[205]

FIG. 2.—Maya Phonetic Terminals.
]

The upper figure he reads _kinil_, the lower _cim-il_. The two signs are the title to a picture in the Codex Troano representing a storm with destruction of human life. The two words _kin-il cim-il_ maybe translated “At the time of the killing.” The syllable _cim_ is expressed in several variants in the Codices, examples of two of which, from the Dresden Codex, are presented in Fig. 3.

FIG. 3.—Maya Phonetic Terminals.
]

The signs for the four cardinal points appear to be expressed phonetically. They are represented in Figs. 4 and 5. The words are for North, _xaman_, East, _lakin_, South, _nohil_, West, _chikin_. Of these the syllable _kin_ appears in _lakin_ and _chikin_, and is represented as above described. The word for North has not been analyzed; that for South has been translated by Prof. Londe Rosny as _ma ya_, the word _ma_ meaning hands or arms, the lower as either a fruit or the masculine sign, in either case the phonetic value being alone intended. Both the name and the etymology are, however, doubtful, resting upon late and imperfect authorities.

By pursuing the plan here indicated, that is, by assuming that a figure whose representative value is known, has also a merely phonetic value in other combinations, a certain number of phonetic elements of the Maya tongue have been identified. Prof. Cyrus Thomas, in an article published in one of our prominent journals, states that he has “interpreted satisfactorily to himself twelve or fifteen compound characters which appear to be phonetic.”[206]

FIGS. 4 and 5.—Signs of the Cardinal Points in Maya.
]

It is obvious, however, that small progress has been made in this direction compared to the labor expended. By far the greater number of the fixed symbols of the Maya are yet undeciphered. It is acknowledged by all recent students that they cannot be representative, as they recur too frequently. To explain them, there is but one sure course, and that is, by a close analysis of the Maya language to get at the relations of ideas in the native mind as expressed in their own phonetic system.

When we turn to the Mexican system of writing, much more definite and extensive information as to its phonetic elements awaits us. It is possible that at bottom it has really no higher phonetic character, but several facts have combined to give us a better understanding of its structure. In the first place, more examples of it have been preserved, some of these with more or less accurate translations. Again, the earlier writers, those whom we look upon as our historical authorities, have been more explicit and ample in their description of Mexican native literature than of that of Yucatan. Finally, and most important, the Mexican language, the Nahuatl, was studied at an early date, and with surprising thoroughness, by the Catholic priests. Within a generation after the conquest they had completed a quite accurate analysis of its grammatical structure, and had printed a Nahuatl-Spanish dictionary containing more words than are to be found in any English dictionary for a century later.

These intelligent missionaries acquainted themselves with the principles of the Mexican script, and to a limited extent made use of it in their religious instructions, as did also the Spanish scriveners in their legal documents in transactions with the natives. They found the native phonetic writing partly syllabic and partly alphabetic; and it was easy for the priests to devise a wholly alphabetic script on the same plan. An interesting example of this is preserved in the work of Valades, entitled _Rhetorica Christiana_, written about 1570. Familiar objects are represented, chiefly of European introduction. Each has the phonetic value only of the first letter of its Nahuatl name. The plan is extremely simple, and indeed the forms and names of the Hebrew letters seem to indicate that they arose in the same way. Applying it to English, we should spell the word _cat_ by a picture of a chair, of an axe, and of a table, each of these being the recognized symbol of its first phonetic element or initial letter. Often any one of several objects whose names begin with the same letter could be used, at choice. This is also illustrated in Valades’ alphabet, where, for instance, the letter _E_ is represented by four different objects.

As I have observed, the native genius had not arrived at a complete analysis of the phonetic elements of the language; but it was distinctly progressing in that direction. Of the five vowels and fourteen consonants which make up the Nahuatl alphabet, three vowels certainly, and probably three consonants, had reached the stage where they were often expressed as simple letters by the method above described. The vowels were _a_, for which the sign was _atl_, water; _e_ represented by a bean, _etl_; and _o_ by a footprint, or path, _otli_; the consonants were _p_, represented either by a flag, _pan_, or a mat, _petl_; _t_, by a stone, _tetl_, or lips, _tentli_; and _z_, by a lancet, _zo_. These are, however, exceptions. Most of the Nahuatl phonetics were syllabic, sometimes one, sometimes two syllables of the name of the object being employed. When the whole name of an object or most of it was used as a phonetic value, the script remains truly phonetic, but becomes of the nature of a rebus, and this is the character of most of the phonetic Mexican writing.

Every one is familiar with the principle of the rebus. It is where a phrase is represented by pictures of objects whose names bear some resemblance in sound to the words employed. A stock example is that of the gallant who to testify his devotion to the lady of his heart, whose name was Rose Hill, had embroidered on his gown the pictures of a rose, a hill, an eye, a loaf of bread, and a well, which was to be interpreted, “Rose Hill I love well.”

In medieval heraldry this system was in extensive use. Armorial bearings were selected, the names of the elements of which expressed that of the family who bore them. Thus Pope Adrian IV, whose name was Nicolas Breakespeare, carried the device of a spear with a broken shaft; the Boltons of England wear arms representing a cask or _tun_ pierced by a cross-bow shaft or _bolt_; etc. Such arms were called _canting_ arms, the term being derived from the Latin _cantare_, to sing or chant, the arms themselves chanting or announcing the family surname.

We have, so far as I am aware, no scientific term to express this manner of phonetic writing, and I propose for it therefore the adjective _ikonomatic_, from the Greek _eikon_, a figure or image, and _onoma_ (genitive, _onomatos_) name,—a writing by means of the names of the figures or images represented. The corresponding noun would be _ikonomatography_. It differs radically from picture-writing (_Bilderschrift_,) for although it is composed of pictures, these were used solely with reference to the sound of their names, not their objective significance.

FIG. 6.—Mexican Phonetic Hieroglyphics of the name of Montezuma.
]

The Mexicans, in their phonetic writing, were never far removed from this ikonomatic stage of development. They combined, however, with it certain clearly defined monosyllabic signs, and the separate alphabetic elements which I have already noted. An examination of the MSS. proves that there was no special disposition of the parts of a word. In other words, they might be arranged from right to left or from left to right, from below upwards or from above downwards; or the one may be placed within the other. It will easily be seen that this greatly increases the difficulty of deciphering these figures.

As illustrations of the phoneticism of Mexican writing I show two compounds, quoted by M. Aubin in his well-known essay on the subject. The first is a proper noun, that of the emperor Montezuma (Fig. 6). It should be read from right to left. The picture at the right represents a mouse trap, in Nahuatl, _montli_, with the phonetic value _mo_, or _mon_; the head of the eagle has the value _quauh_, from _quauhtli_; it is transfixed with a lancet, _zo_; and surmounted with a hand, _maitl_, whose phonetic value is _ma_; and these values combined give _mo-quauh-zo-ma_.

FIG. 7.—Mexican Phonetic Hieroglyphics of the name of a Serpent.
]

The second example is a common noun, the name of a serpent _tecuhtlacozauhqui_ (Fig. 7). It is also read from right to left; the head with the peculiar band and frontal ornament is that of one of the noble class, _tecuhtli_; at the base of the left figure is a familiar sign for _tla_, and represents two teeth, _tlantli_; they are surmounted by a jar, _comitl_ with the value _co_; and this in turn is pierced by a lancet, which here has only its alphabetic value _z_. The remainder of the word was not expressed in the writing, the above signs being deemed sufficient to convey the idea to the reader.

In presenting these examples I do not bring forward anything new. They are from an essay which has been in print nearly forty years.[207] Many other examples are to be seen in the great work of Lord Kingsborough, and later in publications in the city of Mexico. The learned Ramirez undertook a dictionary of Nahuatl hieroglyphics which has in part been published; Orozco y Berra in his “History of Ancient Mexico” gathered a great many facts illustrative of the phonetic character of the Mexican script; and within a year Dr. Peñafiel has issued a quarto of considerable size giving ancient local Mexican names with their phonetic representations.[208]

With these aids at command, why has not our progress in the interpretation of the ancient records on stone and paper been more rapid? Why do we stand now almost at the same point as in 1850?

There can be but one answer, and that will immediately suggest itself from the nature of the phoneticism in the Mexican writing. What I have called the _ikonomatic_ system of writing can be elucidated only by one who has a wide command of the vocabulary of the language. Consider, for a moment, the difficulty which we experience, with all our knowledge of our native tongue, in solving one of the rebuses which appear in the puzzle columns of periodicals for children; or in interpreting the canting arms in armorial bearings. Not only must we recall the various names of the objects represented, and select from them such as the sense of the context requires, but we must make allowance for extensive omissions, as in one of the examples above quoted (Fig. 7), and for mere similarities of sound, often quite remote, as well as for the abbreviations and conventionalisms of practiced scribes, familiar with their subject and with this method of writing the sounds of their language.

Such difficulties as these can only be overcome by long-continued application to the tongues themselves, and by acquainting one’s self intimately with the forms, the methods, and the variations of this truly puzzling graphic system. Every identification is solving an enigma; but once solved, each illustrates the method, confirms its accuracy, and facilitates the learner’s progress, and at the same time stimulates him with the joyous sense of difficulties conquered, and with the vision of discovered truth illuminating his onward path.

Although, as I have stated, the general principles of this method were pointed out forty years ago, the prevailing ignorance of the Nahuatl language has prevented any one from successfully deciphering the Mexican script. This ignorance has had even a worse effect. Men who did not know a dozen words of Nahuatl, who were unable to construe a single sentence in the language, have taken upon themselves to condemn Aubin’s explanations as visionary and untrue, and to deny wholly the phonetic elements of the Mexican writing. Lacking the essential condition of testing the accuracy of the statement, they have presumed blankly to condemn it!

THE IKONOMATIC METHOD OF PHONETIC WRITING.[209]

All methods of recording ideas have been divided into two classes, Thought Writing and Sound Writing.

The first, simplest and oldest is Thought Writing. This in turn is subdivided into two forms, Ikonographic and Symbolic Writing. The former is also known as Imitative, Representative or Picture Writing. The object to be held in memory is represented by its picture, drawn with such skill, or lack of skill, as the writer may possess. In Symbolic Writing, a single characteristic part or trait serves to represent the whole object; thus, the track of an animal will stand for the animal itself; a representation of the peculiar round impression of the wolf’s foot, or the three-lined track of the wild turkey, being amply sufficient to designate these creatures. Even the rudest savages practice both these forms of writing, and make use of them to scratch on rocks, and paint on bark and hides, the record of their deeds.

It will be observed that Thought Writing has no reference to spoken language; neither the picture of a wolf, nor the representation of his footprint, conveys the slightest notion of the sound of the word _wolf_. How was the enormous leap made from the thought to the sound—in other words, from an ideographic to a phonetic method of writing?

This question has received considerable attention from scholars with reference to the development of the two most important alphabets of the world, the Egyptian and the Chinese. Both these began as simple picture writing, and both progressed to almost complete phoneticism. In both cases, however, the earliest steps are lost, and can be retraced only by indications remaining after a high degree of phonetic power had been reached. On the other hand, in the Mexican and probably in the Maya hieroglyphics, we find a method of writing which is intermediate between the two great classes I have mentioned, and which illustrates in a striking manner the phases through which both the Egyptian and Semitic alphabets passed somewhat before the dawn of history.

To this method, which stands midway between the ikonographic and the alphabetic methods of writing, I have given the name _ikonomatic_, derived from the Greek εικων-ονος, an image, a figure; ονομα-ατος, a name. That which the figure or picture refers to is not the object represented, but the _name_ of that object—a _sound_, not a _thing_. But it does not refer to that sound as the name of the object, but precisely the contrary—it is the sound of the name of some other object or idea. Many ideas have no objective representation, and others are much more simply expressed by the use of figures whose names are familiar and of similar sound. Thus, to give a simple example, the infinitive “to hide” could be written by a figure 2, and the picture of a skin or hide. It is this plan on which those familiar puzzles are constructed which are called _rebuses_, and none other than this which served to bridge over the wide gap between Thought and Sound writing. It is, however, not correct to say that it is a writing by _things_, “_rebus_;” but it is by the _names_ of things, and hence I have coined the word _ikonomatic_, to express this clearly.

I shall select several illustrations from two widely diverse sources, the one the hieroglyphs of Egypt, the other the heraldry of the Middle Ages, and from these more familiar fields obtain some hints of service in unraveling the intricacies of the Mexican and Maya scrolls.

The general principle which underlies “ikonomatic writing” is the presence in a language of words of different meaning but with the same or similar sounds; that is, of _homophonous_ words. The figure which represents one of these is used phonetically to signify the other. There are homophones in all languages; but they abound in some more than in others. For obvious reasons, they are more abundant in languages which tend toward monosyllabism, such as the Chinese and the Maya, and in a less degree the ancient Coptic. In these it is no uncommon occurrence to find four or five quite different meanings to the same word; that is, the same sound has served as the radical for that many different names of diverse objects. The picture of any of these objects would, to the speaker of the language, recall a sound which would have all these significations, and could be employed indifferently for any of them. This circle of meanings would be still more widely extended when mere similarity, not strict identity, was aimed at.

Such was plainly the origin of phoneticism in the Egyptian hieroglyphic inscriptions. Take the word _nefer_. Its most common concrete signification was “a lute,” and in the picture writing proper the lute is represented by its figure. But _nefer_ had several other significations in Coptic. It meant, a _colt_, a _conscript soldier_, a _door_, and the adjective _good_. The picture of the lute therefore was used to signify every one of these.

It will be observed that this is an example of a pure ikonograph—the picture is that of the object in full, a lute; but precisely in the same way the second class of figures in picture writing, those which are wholly symbolic, may be employed. This, too, finds ample illustration in the Egyptian hieroglyphics. Instead of the picture of a house, the figure of a square was employed, with one side incomplete. Phonetically, this conveyed the sound _per_, which means _house_, and several other things.

It will readily be seen that where a figure represents a number of homophonous words, considerable confusion may result from the difficulty of ascertaining which of these is intended. To meet this, we find both in Egyptian and Chinese writing series of signs which are written but not pronounced, called “determinatives.” These indicate the class to which a word has reference. They are ideographic, and of fixed meaning. Thus, after the word _nefer_, when used for conscript, the determinative is the picture of a man, etc.[210]

There is little doubt but that all the Egyptian syllabic and alphabetic writing was derived from this early phase, where the governing principle was that of the rebus. At the date of the earliest inscriptions, most of the phonetics were monosyllabic; but in several instances, as _nefer_, above given, _neter_, which represents a banner, and by homophony, a god, and others, the full disyllabic name was preserved to the latest times. The monosyllabic signs were derived from the initial and the accented syllables of the homophones; and the alphabet, so-called, but never recognized as such, by the Egyptians, either from monoliteral words, or from initial sounds. At no period of ancient Egyptian history was one sound constantly represented by one sign. In the so-called Egyptian alphabet, there are four quite different signs for the _M_, four for the _T_, three for the _N_, and so on. This is obviously owing to the independent derivation of these phonetic elements from different figures employed ikonomatically.

There are other peculiarities in the Egyptian script, which are to be explained by the same historic reason. For instance, certain phonetic signs can be used only in definite combinations; others must be assigned fixed positions, as at the beginning or at the end of a group; and, in other cases, two or more different signs, with the same phonetic value, follow one another, the scribe thinking that if the reader was not acquainted with one, he would be with the other. I note these peculiarities, because they may be expected to recur in other systems of ikonomatic writing, and may serve as hints in interpreting them.

Evidently, one of the earliest stimuli to the development of phonetics was the wish to record proper names, which in themselves had no definite signification, such as those drawn from a foreign language, or those which had lost through time their original sense. In savage conditions every proper name is significant; but in conditions of social life, as developed as that of the Egyptians of the earlier dynasties, and as that of the Mayas and Mexicans in the New World, there are found many names without meaning in the current tongue. These could not be represented by any mode of picture writing. To be recorded at all, they must be written phonetically; and to accomplish this the most obvious plan was to select objects whose names had a similar sound, and by portraying the latter, represent to the ear the former. The Greek names, _Alexander_ and _Alexandria_, occurring on the Rosetta Stone, were wholly meaningless to the Egyptian ear; but their scribes succeeded in expressing them very nearly by a series of signs which in origin are rebuses.

This inception of the ikonomatic method, in the effort to express phonetically proper names, is admirably illustrated in mediæval heraldry. Very early in the history of armorial bearings, we find a class of scutal devices called in Latin _arma cantantia_, in English _canting arms_, in French _armes parlantes_. The English term _canting_ is from the Latin _cantare_, in its later sense of _chanting_ or _announcing_. Armorial bearings of this character present charges, the names of which resemble more or less closely in sound the proper names of the family who carry them.

Some writers on heraldry have asserted that bearings of this character should be considered as what are known as _assumptive arms_, those which have been _assumed_ by families, without just title. Excellent authorities, however, such as Woodham and Lower, have shown that these devices were frequent in the remotest ages of heraldry.[211] For instance, in the earliest English Roll of Arms extant, recorded in the reign of the third Henry, about the year 1240, nine such charges occur, and still more in the Rolls of the time of Edward the Second. They are also abundant in the heraldry of Spain, of Italy and of Sweden; and analogous examples have been adduced from ancient Rome. In fact, the plan is so obvious that instances could be quoted from every quarter of the globe. In later centuries, such punning allusions to proper names became unpopular in heraldry, and are now considered in bad taste.

To illustrate their character, I will mention a few which are of ancient date. The well-known English family of _Dobells_ carry a _hart passant_, and three bells _argent_, thus expressing very accurately their name, _doe-bells_. The equally ancient family of Boltons carry a device representing a cask or _tun_, transfixed by a cross-bow or _bolt_. Few canting arms, however, are so perfect as these. The Swinburnes, who are among those mentioned on the Roll of 1240, already referred to, bear three boar-heads, symbolical of _swine_; the Boleynes carry three bulls’ heads, which reminds us of Cardinal Wolsey’s pronunciation of the name in Shakespeare’s Henry VIII, _Bullen_:

“Anne Bullen? No; I’ll no Anne Bullens for him;
There’s more in’t than fair visage.—Bullen!
No, we’ll no Bullens.”—_King Henry VIII, Act III._

Not rarely the antiquity of such bearings is evidenced by the loss of the allusion in the current language, and recourse must be had to ancient and obsolete words to appreciate it. The English Harrisons display in their shield a hedge-hog, which is to be explained by the French _hérisson_, and testifies to their Norman origin. The Sykes of the north of England show a fountain in their shield, whose significance is first ascertained on learning that in the Northumbrian dialect _syke_ means a flowing spring or stream. The celebrated _fleurs-de-lys_ of the royal house of France are traced back to the first Louis, whose name was pronounced _Loys_, and from the similarity of this to the common name of the flower, the latter was adopted as the charge on his shield.

Hundreds of such examples could be adduced, and the task of examining and analyzing them would not be an altogether vain one, as the principles upon which they were applied are the same which control the development of ikonomatic writing wherever we find it. But I pass from the consideration of these facts of general knowledge to the less known and much misunderstood forms of this writing which are presented in American archæology.

These are best exemplified in the so-called Mexican picture writing. For many years scholars have been divided in opinion whether this was purely ikonographic or partly phonetic. About forty years ago M. Aubin wrote an essay maintaining that it is chiefly phonetic, and laid down rules for its interpretation on this theory. But neither he nor any who undertook to apply his teachings succeeded in offering any acceptable renderings of the Aztec Codices. I am persuaded, however, that the cause of this failure lay, not in the theory of Aubin, but in the two facts, first, that not one of the students who approached this subject was well grounded in the Nahuatl language; and, secondly, that the principles of the interpretation of ikonomatic writing have never been carefully defined, and are extremely difficult, ambiguous and obscure, enough so to discourage any one not specially gifted in the solution of enigmas. At first, every identification is as puzzling as the effort to decipher an artificial rebus.

There are, indeed, some able scholars who still deny that any such phoneticism is to be found in Mexican pictography. To convince such of their error, and to illustrate the methods employed by these native American scribes, I will present and analyze several typical examples from Aztec manuscripts.

Beginning with proper names drawn from other languages, we find that the Nahuas had a number of such, which, of course, had no meaning in their own tongue. One of their documents speaks of the town of the Huastecas, called by that tribe _Tamuch_, which means in their tongue “near the scorpions,” and by the Aztecs, in imitation, _Tamuoc_.[212] As the Huasteca is a Maya dialect, totally distinct from the Nahuatl, this word had no sense to the ears of the Aztecs. To convey its sound, they portrayed a man holding in his hands a measuring stick, and in the act of measuring. Now, in Nahuatl, the verb “to measure” is _tamachina_; the measuring stick is _octocatl_; and to make the latter plainer, several foot-prints, _xoctli_, are painted upon the measuring stick, giving an example of the repetition of the sound, such as we have already seen was common among the Egyptian scribes.

FIG. 1.—Tamuoc.
]

FIG. 2.—Mapachtepec.
]

In another class of proper names, in their own tongue, although they had a meaning in the Nahuatl, the scribe preferred to express them by ikonomatic instead of ikonographic devices. Thus, _Mapachtepec_, means literally, “badger hill,” or “badger town,” but in place of depicting a badger, the native writer made a drawing of a hand grasping a bunch of Spanish moss, the _Tillandsia usneoides_. The hand or arm in Nahuatl is _maitl_, the moss _pachtli_; and taking the first syllables of these two words we obtain _ma pach_: the word _tepec_, locative form of _tepetl_, hill or village, is expressed by the usual conventional ideographic or determinative sign.

In other names, the relative _positions_ of the objects are significant, reminding us of the rebus of a well-known town in Massachusetts, celebrated for its educational institutions:

&
Mass.

which is to be read, “Andover, Massachusetts;” so in the Aztec scrolls, we have _itzmiquilpan_ represented by an obsidian knife, _itztli_, and an edible plant, _quilitl_, which are placed above or over (_pan_), the sign for cultivated land, _milli_, thus giving all the elements of the name, the last syllable by position only.

FIG. 3.—Itzmiquilpan.
]

In one respect I believe the ikonomatic writing of the Mexicans is peculiar; that is, in the phonetic value which it assigns to _colors_. Like the Egyptian, it is polychromatic, but, so far as I know, the Egyptian polychromes never had a phonetic value; they were, in a general way, used by that people as determinatives, from some supposed similarity of hue; thus green indicates a vegetable substance or bronze, yellow, certain woods and some animals, and so on. In heraldry the colors are very important and have well-defined significations, but very seldom, if ever, phonetic ones. Quite the contrary is the case with the Mexican script. It presents abundant instances where the color of the object as portrayed is an integral phonetic element of the sound designed to be conveyed.

To quote examples, the Nahuatl word for yellow is _cuztic_ or _coztic_, and when the hieroglyphics express phonetically such proper names as _Acozpa_, _Cozamaloapan_, _Cozhuipilcan_, etc., the monosyllable _coz_ is expressed solely by the yellow color which the scribe lays upon his picture. Again, the name _Xiuhuacan_, “the place of grass,” is represented by a circle colored pale blue, _xiuhtic_. The name of this tint supplies the phonetic desired. The name of the village _Tlapan_ is conveyed by a circle, whose interior is painted red, _tlapalli_, containing the mark of a human foot-print. Such examples are sufficient to prove that in undertaking to decipher the Mexican writing we must regard the color as well as the figure, and be prepared to allow to each a definite phonetic value.

FIG. 4.—Acozpa. (A yellow center surrounded by water drops, _atl_,
_a_.)
]

FIG. 5.—Tlamapa.
]

It must not be understood that all the Aztec writing is made up of phonetic symbols. This is far from being the case. We discover among the hundreds of curious figures which it presents, determinatives, as in the Egyptian inscriptions, and numerous ideograms. Sometimes the ideogram is associated with the phonetic symbol, acting as a sort of determinative to the latter. An interesting example of this is given at the beginning of the “Manuscrito Hieratico,” recently published by the Spanish government.[213] It is the more valuable as an example, as the picture writing is translated into Nahuatl and written in Spanish characters. The date of the document, 1526, leaves no doubt that it is in the same style as the ancient Codices. The page is headed with the picture of a church edifice; underneath is the outline of a human arm, and the legend in Nahuatl is:

_In Altepetl y Santa Cruz Tlamapa._

These words mean, “the town of Santa Cruz Tlamapa.” The name “_tlamapa_” means “on the hill-side,” and doubtless originally referred to the position in which the village was situated. But the prefix “_tlama_” usually signifies, “to do something with the arms or hands,” derived from _maitl_, hand or arm. Hence, the figure of the extended arm gives this disyllable, _tlama_, which was sufficient to recall the name of the town.

The Aztecs by no means confined the ikonomatic system to proper names. They composed in it words, sentences, and treatises on various subjects. In proportion as it is applied to these connected and lengthy compositions, its processes become more recondite, curious and difficult of interpretation. Without a knowledge of the spoken language considerably more than rudimentary, it would be hopeless for the student to attempt to solve the enigmas which he meets at every step. Yet every well-directed effort will convince him that he is on the right track, and he will constantly be cheered and stimulated to further endeavor by the victories he will win day by day.

The analogy which is presented in so many particulars between Mexican and Maya civilization would lead us to infer that the Maya writing, of which we have a number of examples well preserved, should be unlocked by the same key which has been successfully applied to the Aztec Codices. The latest writers on the Maya manuscripts, while agreeing that they are in part, at least, in phonetic characters, consider them mostly ideographic. But it is to be noted that not one of these writers had any practical acquaintance with the sounds of the Maya language, and scarcely any with its vocabulary. From this it is evident that even were these codices in ikonomatic writing, such investigators could make very little progress in deciphering them, and might readily come to the conclusion that the figures are not phonetic in any sense. Precisely the same position was taken by a number of students of Egyptian antiquity long after the announcement of the discovery of Champollion; and even within a few years works have been printed denying all phoneticism to the Nilotic inscriptions.

What induces me to believe that much of the Maya script is of the nature of the Mexican is the endeavor, undertaken for a very different purpose, of Professor Valentini to explain the origin of the so-called Maya alphabet, preserved by Bishop Landa, and printed in the editions of his celebrated “Description of Yucatan.”[214] Professor Valentini shows by arguments and illustrations, which I think are in the main correct, that when the natives were asked to represent the sounds of the Spanish letters in their method of writing, they selected objects to depict, whose names, or initial sounds, or first syllables, were the same, or akin, to the sounds of the Spanish vowel or consonant heard by them. Sometimes they would give several words, with their corresponding pictures, for the same sound; just as I have shown was the custom of the ancient Egyptians. Thus, for the sound _b_ they drew a foot-print, which in their tongue was called _be_; for the sound _a_ an obsidian knife, in Maya, _ach_, etc. Valentini thinks also that the letter _e_ was delineated by black spots, in Maya _eek_, meaning black, which, if proved by further research, would show that the Mayas, like the Mexicans, attributed phonetic values to the colors they employed in their painted scrolls.

Outside of the two nations mentioned, the natives of the American continent made little advance toward a phonetic system. We have no positive evidence that even the cultivated Tarascas and Zapotecs had anything better than ikonographs; and of the Quiches and Cakchiquels, both near relatives of the Mayas, we only know that they had a written literature of considerable extent, but of the plan by which it was preserved we have only obscure hints. Next to these we should probably place the Chipeway pictography, as preserved on their _meda_ sticks, bark records, and _adjidjiatig_ or grave-posts. I have examined a number of specimens of these, but have failed to find any evidence that the characters refer to sounds in the language; however, I might not consider it improbable that further researches might disclose some germs of the ikonomatic method of writing even in these primitive examples of the desire of the human intellect to perpetuate its acquisitions, and hand them down to generations yet unborn.

THE WRITING AND RECORDS OF THE ANCIENT MAYAS.[215]

_1.—Introductory._

One of the ablest living ethnologists has classified the means of recording knowledge under two general headings—Thought-writing and Sound-writing.[216] The former is again divided into two forms, the first and earliest of which is by pictures, the second by picture-writing.

The superiority of picture-writing over the mere depicting of an occurrence is that it analyzes the thought and expresses separately its component parts, whereas the picture presents it as a whole. The representations familiar among the North American Indians are usually only pictures, while most of the records of the Aztec communities are in picture-writing.

The genealogical development of Sound-writing begins by the substitution of the sign of one idea for that of another whose sound is nearly or quite the same. Such was the early graphic system of Egypt, and such substantially to-day is that of the Chinese. Above stands syllabic writing, this as that of the Japanese, and the semi-syllabic signs of the old Semitic alphabet; while, as the perfected result of these various attempts, we reach at last the invention of a true alphabet, in which a definite figure corresponds to a definite elementary sound.

It is a primary question in American archæology, How far did the most cultivated nations of the western continent ascend this scale of graphic development? This question is as yet unanswered. All agree, however, that the highest evolution took place among the Nahuatl-speaking tribes of Mexico and the Maya race of Yucatan.

I do not go too far in saying that it is proved that the Aztecs used to a certain extent a phonetic system of writing, one in which the figures refer not to the thought, but to the sound of the thought as expressed in spoken language. This has been demonstrated by the researches of M. Aubin, and, of late, by the studies of Señor Orozco y Berra.[217]

Two evolutionary steps can be distinguished in the Aztec writing. In the earlier the plan is that of the rebus in combination with ideograms, which latter are nothing more than the elements of picture-writing. Examples of this plan are the familiar “tribute rolls” and the names of towns and kings, as shown in several of the codices published by Lord Kingsborough. The second step is where a conventional image is employed to represent the sound of its first syllable. This advances actually to the level of the syllabic alphabet; but it is doubtful if there are any Aztec records entirely, or even largely, in this form of writing. They had only reached the commencement of its development.

The graphic system of the Mayas of Yucatan was very different from that of the Aztecs. No one at all familiar with the two could fail at once to distinguish between the manuscripts of the two nations. They are plainly independent developments.

We know much more about the ancient civilization of Mexico than of Yucatan; we have many more Aztec than Maya manuscripts, and hence we are more at a loss to speak with positiveness about the Maya system of writing than about the Mexican. We must depend on the brief and unsatisfactory statements of the early Spanish writers, and on what little modern research has accomplished, for means to form a correct opinion; and there is at present a justifiable discrepancy of opinion about it among those who have given the subject most attention.

_2.—Descriptions by Spanish Writers._

The earliest exploration of the coast of Yucatan was that of Francisco Hernandez de Cordova, in 1517. The year following a second expedition, under Juan de Grijalva, visited a number of points between the island of Cozumel and the Bahia de Terminos.

Several accounts of Grijalva’s voyage have been preserved, but they make no distinct reference to the method of writing they found in use. Some native books were obtained, however, probably from the Mayas, and were sent to Spain, where they were seen by the historian Peter Martyr. He describes them in general terms, and compares the characters in which they were written to the Egyptian hieroglyphics, some of which he had seen in Rome. He supposes that they contain the laws and ceremonies of the people, astronomical calculations, the deeds of their kings, and other events of their history. He also speaks in commendation of the neatness of their general appearance, the skill with which the drawing and painting were carried out. He further mentions that the natives used this method of writing or drawing in the affairs of common life.[218]

Although Yucatan became thus early known to the Spaniards, it was not until 1541 that a permanent settlement was effected, in which year Francisco de Montejo, the younger, advanced into the central province of Ceh Pech, and established a city on the site of the ancient town called _Ichcanziho_, which means “the five (temples) of many oracles (or serpents),” to which he gave the name _Mérida_, on account of the magnificent ancient edifices he found there.

Previous to this date, however, in 1534, Father Jacobo de Testera, with four other missionaries, proceeded from Tabasco up the west coast to the neighborhood of the Bay of Campeachy. They were received amicably by the natives, and instructed them in the articles of the Christian faith. They also obtained from the chiefs a submission to the King of Spain; and I mention this early missionary expedition for the fact stated that each chief signed this act of submission “with a certain mark, like an autograph.” This document was subsequently taken to Spain by the celebrated Bishop Las Casas.[219] It is clear from the account that some definite form of signature was at that time in use among the chiefs.

It might be objected that these signatures were nothing more than rude totem marks, such as were found even among the hunting tribes of the Northern Mississippi Valley. But Las Casas himself, in whose possession the documents were, here comes to our aid to refute this opinion. He was familiar with the picture-writing of Mexico, and recognized in the hieroglyphics of the Mayas something different and superior. He says expressly that these had inscriptions, writings, in certain characters, the like of which were found nowhere else.[220]

One of the early visitors to Yucatan after the conquest was the Pope’s commissary-general, Father Alonzo Ponce, who was there in 1588. Many natives who had grown to adult years in heathenism must have been living then. He makes the following interesting observation: