CC BY
0
Егорова М.А.
Кандидат политических наук, доцент кафедры иностранных языков факультета гуманитарных и социальных наук (ФГСН). Российский университет дружбы народов имени Патриса Лумумбы.
Егоров А.А.
Д.ф.-м.н., внештатный профессор-консультант. Российский университет дружбы народов имени Патриса Лумумбы.
Идентификация иероглифов в древних надписях
PhD of Political Sciences, Associate Professor, Department of Foreign Languages, Faculty of Humanities and Social Sciences. Рeoples' Friendship University of Russia.
Identification of the hieroglyphs in the ancient inscriptions
Jiaguwen (jiaguwen / XIV-XI centuries BC) inscriptions
on turtle shells, oracle bones and bronze vessels Jinwen (jinwen / the earliest of which date back to the end of 2nd millennium BC) belong to the most ancient samples of Chinese history, archeology, linguistics and science [1-12]. Basically, their research is devoted to the study of hieroglyphic inscriptions in order to interpret their content, as well as identify ancient keys or Chinese radicals (bushou / p^^).
This article uses a new research method [5, 11, 12], which can be a good addition to the traditional methods of (mainly visual) research used now [1-4, 11]. Using digital data from photometry of the artifact relief, this method allows determining various characteristics and parameters, including statistical ones, characterizing the properties of the studied ancient surface [5, 12]. The objects under study are the oldest Chinese artifacts such as Jiaguwen and Jinwen Various artifacts were examined, but the main results are given only for Jiaguwen. This Jiaguwen
Egorova M.A.
Egorov A.A.
Dr.Sc., Visiting Professor-consultant. Рeoples' Friendship University of Russia.
is shown in Fig.1 [3, 5, 12] (The image used in this paper was taken from open sources and is used as an illustration.).
Main attention is paid to the development of new methods for the correct identification of the same type of hieroglyphic signs in the hieroglyphic inscriptions of Jiaguwen according to the characteristic digitized data. The solution to this important problem will make it possible to avoid mistakes at the initial stage, when the correct recognition of hieroglyphic signs is carried out according to the obtained digital profiles. As a result, at all subsequent stages it becomes possible to analyze the correct data. Recall that our method is based on photometry of surface profiles, in which a sample of brightness levels in a discrete set of points is carried out with subsequent conversion to digital form [5, 12-14]. Photometric data of the surface profile were obtained using "Fiji ImageJ" program for Windows. The obtained data are digitally processed in order to determine the characteristics and parameters characterizing the properties of the researched surfaces. As an example, you can see the results of photometry of two separated hieroglyphs [12]: wang - network, grid Й, and yue ^ - moon, month; meat. In particular, it can also be the most important profile's statistical characteristics [5].
Fig. 1 shows a photograph of one of the studied samples (tortoise shell). In the photo, the numbers "1" and "2": mark two of the same type of ancient hieroglyphs (hereinafter — hieroglyphs "1", "2") of the type "home, family" (^). The inset in Fig. 1 at the bottom right under the ruler shows an enlarged hieroglyph "2" (marked as "3"). Ruler allows one to accurately determining the linear dimensions of the hieroglyphs. Numbers "6, 7, 8" in the inset on the right mark horizontal lines, which shows its approximate position relative to the characters "1-4", when the photometry of their brightness profiles were performed.
The development of a method for finding the standard characteristics of the Jiaguwen hieroglyphic inscriptions was carried out in four main stages. All results were obtained for a number of hieroglyphs, e.g., such as: "home, family' (Щ), "stand" (Й), "exchange, transmit, hand over" (^). However, the presentation of the method is illustrated only for the first hieroglyph^.
The proposed method was implemented in four stages (details are given in Appendix). The analysis showed that the function under study is best characterized by adjustable functions that describe harmonic vibrations. The essence of harmonic analysis in this case lies in the representation of the studied functions in the form of the sum of some harmonic oscillations and in the study of the properties of the obtained profile of the studied ancient artifact. We finally find a form of the digitized profile, which characterizes the studied ancient hieroglyph "home, family" on average. The resulting set of characteristics and parameters can be considered as
some standard for this hieroglyph. Similarly, one can get a standard for other hieroglyphs.
All this indicates a certain universality of the method described in this article for studying various ancient hieroglyphs, for example, Chinese hieroglyphs or Chinese radicals (bush o u) in the Jiaguwen (jiaguwen) on turtle shells, and Jinwen (jinwen) on bronze vessels, the earliest of which date back to the end of the 2nd millennium BC. The results obtained allow us to state that for correct identification of the same type of signs in the hieroglyphic inscriptions of Jiaguwen and Jinwen according to typical (standard) data, one can apply methods of harmonic analysis, e.g., the method of Fourier analysis. However, if it is necessary, more complex theories and appropriate calculation algorithms can be used (see e.g. [5, 12-14]).
Definitely, Oracle scripts played an important role in the process of the dissemination of the ancient Chinese language throughout China. Thus, Chinese language was gradually introduced into those areas where it was historically not widespread, especially during V-th century BC - Il-nd century AD [1-9, 11].
When solving the problem of finding the number of hieroglyphs using for example the numerical Fourier transform, we have shown previously the possibility of not only a stable, but also an unambiguous solution to the problem of determining the quantity of hieroglyphic signs, as well as their location.
The proposed
methods cannot com-
Fig. 1. Photo of the test sample Jiaguwen on which some of the studied hieroglyphs are marked: "1", "2" are "home, family" (^), "3", "4" are "stand" (^), "5" is "exchange, transfer, hand over" (^). The number "6" mark the horizontal lines along which the photometry was carried out.
Fig. 2. Digital 3D inverted image of hieroglyph "home, family" (see Fig. 1).
pletely replace traditional research methods using in history, archeology, and linguistics; however, they undoubtedly allow looking at this scientific problem from new, original positions and obtaining a number of perspective promising results. In particular, this is the possibility of identifying on the surface of ancient samples of hieroglyphic symbols that are not available for research by traditional methods of visual observation.
To test the method, samples of onyx, jasper and jade were used, which surface is similar to the surface of ancient artifacts [5, 12]. We should outline that the photometric data of the surface profile were obtained and processed using different programs for Windows, e.g. "Fiji ImageJ" (see [5, 12]). It's important that original images can be downloaded in raster and vector formats. This gives extra opportunities in further processing of information (see e.g. [5, 12-14]).
As a result, the development of different methods of research and classification of similar elements of the same type of surface structure will make it possible to reliably study ancient poorly preserved samples on which inscriptions are poorly distinguishable by the eye. Undoubtedly, this can contribute to the more productive work of researchers working in these fields of science, art and culture: archaeologists, historians, cultur-ologists, and, of course, linguists [1-13].
Let's give another way of research ancient inscriptions on the oldest artifacts with the help of the computer programs as "Fiji ImageJ" for Windows.
The Fig. 2 shows a digital 3D inverted image of the hieroglyph "2" ("home, family" (Щ)), shown in the Fig. 1 (obtained with the computer program "Fiji ImageJ" for Windows). The hieroglyph is shown in Fig. 2 as in Fig. 1.
One can see how well the original image of the hieroglyph "home, family" has been digitally restored. Even small defects in the original image of the hieroglyph are visible. The use of three-dimensional (3D) reconstruction allows seeing the image in depth, which opens up additional opportunities for identifying the inscriptions on the ancient artifacts under study.
An original method was developed for the study of ancient hieroglyphic inscriptions using the harmonic analysis. The error of the method at this stage of research does not exceed 15-20%. The efficiency of this method is shown on the example of three common types' different hieroglyphs: "home, family" (^), "stand" (Й), "exchange, transmit, hand over' (^). Similar results were obtained for a number of other hieroglyphs.
The results obtained in this paper can be useful not only to further improvement of the similar non-contact optical methods for examining the surface of ancient artifacts, but also, for example, in mathematical linguistics, lexicostatistics and sociolinguistics, especially in the statistical analysis of similar hieroglyphic inscriptions. Therefore, we can talk about
the concept of the symbiosis of culture and science in a broader sense than just the use of scientific achievements in applied quality.
Appendix
At the first stage, let's examine the hieroglyph "1", i.e. first of the hieroglyphs "home, family" Photometric data of the surface profile along the horizontal lines can be obtained using the "Fiji ImageJ" program for Windows. In this case (line with 10 grooves) distance along the horizontal axis equals approximately to 20 cm (or 540 pixels) [5, 12]. The numbers from "1" to "10" in Fig. 1 marked dips in brightness in the curve, corresponding to 10 structural elements of the groove type, some of which are included in the composition of the most ancient hieroglyphs [12]. Computer processing of digital photometric data makes it possible to judge not only about the different structural complexity of the studied hieroglyphs, but also to determine their number and location.
Then various functions were selected that approximated the original and smoothed profiles. At the same time, various procedures for smoothing the original profiles were applied to reduce the effect of interference. In addition, for the convenience of the work, normalized smoothed dependencies were also obtained.
Different types of curves were considered as fitting functions. The analysis showed that the profile under study is best characterized by adjustable functions of the sinusoidal (periodic) types, i.e. the functions that describe harmonic vibrations. The essence of harmonic analysis lies in the representation of investigated functions in the form of the sum of some harmonic oscillations and in the study of properties of the obtained profile.
At the second stage, we examine the character "2" (located to the left of the character "1", see Fig. 1), i.e. second of the hieroglyphs "home, family"
At the third stage, smoothed and normalized brightness dependencies for the characters "1" and "2" "home, family" are jointly studied. Previously, both digitized brightness dependencies were brought to the same level of value at the beginning of the horizontal axis. Then the behavior of, for example, the difference profile was analyzed, when the first dependency for the character "1" was numerically subtracted from the analogous second dependence for the character "2", and vice versa. In this case, for the convenience of analysis, these dependences were shifted slightly relative to each other along the axis.
At the fourth stage, a numerically averaged form of the digitized brightness dependency was obtained, which characterizes the studied ancient hieroglyphs "home, family" on average. For this purpose, we used smoothed and normalized two dependencies for two studied hieroglyphs^. As a result, we got an averaged dependence, providing in the process of photom-
etry along the axis the intersection of only three vertical elements such as solitary grooves. This curve was smoothed again and got a final profile as a result, which can now be seen as one of some of the standard character characteristics of the studied ancient hieroglyphs "home, family" (see Fig. 3).
Fig. 3 shows the smoothed normalized brightness dependence and an adjustable sinusoidal function that can be describe by a period, an amplitude, and a phase (an error is not more than 5-10%). For this averaged final profile we can write the closest approximating sinusoidal function, where the amplitude (about 0.6 arb.u.), period (about 20 arb.u.), initial phase (about 12.3) can be determined with an accuracy of 4 decimals. Moreover, standard error of fitting the parameters does not exceed 17%. Consequently this periodic function with these parameters found using proposed new method can be taken as one of the standard characteristics of the ancient hieroglyph "home, family".
We should note that a typical harmonic vibration of the similar form is characterized by the following parameters: period, oscillation phase, oscillation frequency, speed, and initial phase of oscillation. Best results can be achieved, if to find the average dependence over realizations of the process, i.e. large set of the same type of hieroglyphs obtained during photometry.
Fig. 3 shows one of the obtained smoothed normalized dependences and a fitting sinusoidal function (the error of finding its period is not more than 8%). We emphasize that the resulting averaged smoothed dependence in Fig. 3, characterizes the studied ancient hieroglyphs "home, family" on average. Analogous outcomes were obtained also for amounts of other type of ancient characters. As a result, the main task set in the work was achieved.
All these results indicates a certain universality of the methods described in this article for studying various ancient hieroglyphs, not only Chinese hieroglyphs (or radicals in the oldest artifacts as Jiaguwen and Jinwen), but others too, for example Hittite or Egyptian.
Библиографический список / References
1. Яхонтов С.Е. Древне-китайский язык. - М.: Наука, 1965. Yakhontov S.E. Ancient Chinese language. - M.: Nauka, 1965.
2. Keightley D.N. Sources of Shang history: the oracle-bone inscriptions of Bronze Age China. - London: Berkeley, 1985.
3. High-resolution digital images of oracle bones, Cambridge Digital Library. // URL: http:// cudl.lib.cam.ac.uk/view/MS-CUL-00001-00155/1
4. Egorova M.A., Egorov A.A., Solovieva T.V. Modeling the distribution and modification of writing in proto-Chinese language communities // Automat. Doc. and Math. Ling. 2020. Vol. 54 (2). P. 92-104.
5. Egorova M., Egorov A. The role of ancient written signs in the preservation and development of the Chinese language. Features of the study and analysis of intricately located hieroglyphic inscriptions Jiaguwen // Advances in Social Science, Education and Humanities Research. 2021. Vol. 537. P. 41-47.
Fig. 3. Digitized dependencies (brightness - on the vertical axis, distance - on the horizontal axis) for two hieroglyphs "home, family": the blue curve is the smoothed normalized curve (averaged for both hieroglyphs); the dark cyan curve is the blue curve smoothed again; the red curve is the fitting function of the sinusoidal kind sin(x). 1 arb.u. on the horizontal axis equals to 400 цт.
6. Xueqin L. The Xia-Shang-Zhou chronology project: methodology and results // J. of East Asian Archaeology. 2002. Vol. 4. P. 321-333.
7. The Peopling of East Asia: Putting together archaeology, linguistics and genetics / Eds. Blench R., Sagart L., Sanchez-Mazas A. - London: Routledge Curzon, 2005.
8. Vasil'ev L.S. The origin of ancient Chinese civilization // Voprosy istorii, 1974. № 12. P. 86-102.
9. Gill N.S. Periods and Dynasties of Ancient China. // URL: https://www.thoughtco.com/ periods-and-dynasties-of-ancient-china-117665
10. Jing Z. A brief introduction to new discoveries and research in Chinese archaeology in 2015 // Asian Archaeology. 2018. Vol. 1. P. 129-145.
11. Егорова М.А., Егоров А.А., Соловьёва Т.М. Особенности архаической письменности древнего китайского языка в сравнении с современным: исторический контекст // Вопросы истории. 2021. № 11 (1). С. 189-207.
Egorova M.A., Egorov A.A., Solovyova T.M. Features of the archaic writing of the ancient Chinese language in comparison with the modern one: historical context // Questions of history. 2021. № 11 (1). P. 189-207.
12. Егорова М.А., Егоров А.А., Орлова Т.Г., Трифонова Е.Д. Методы исследования иероглифических надписей на древнейших артефактах - введение в проблему: история, археология, лингвистика // Вопросы истории. 2022. № 3 (1). C. 17-25.
Egorova M.A., Egorov A.A., Orlova T.G., Trifonova E.D. Methods for studying hieroglyphic inscriptions on ancient artifacts - an introduction to the problem: history, archeology, linguistics // Questions of history. 2022. № 3 (1). P. 17-25.
13. Haslam M., Robertson G., Crowther A., Nugent S., Kirkwood L. Archaeological science under a microscope. - ANU Press, 2009. // URL: http://www.jstor.org/stable/jxtt24h7m5
14. Jahne B. Digital image processing. - NY: Springer, 2005.
