1 CAES Vol. 5, № 2 (June 2019)

Comparing ornaments of the Pit-Comb Ware from the site of 1 and the sites of Hepojarvi and Toksovo by Monte Carlo method

Alexander Akulov independent scholar; , ; e-mail: [email protected]

Abstract

The sites of Hepojarvi, Toksovo and Okhta 1 are connected by the Okhta river, so they could belong to the same local group. In order to estimate the degree of proximity of these sites Monte Carlo method of comparing ornamental traditions of pottery was applied to the corresponding sets of potsherds. The degree of proximity of Okhta 1 and Hepojarvi is 0.47; that of Okhta 1 and Toksovo is 0.49. The degree of proximity of Hepojarvi and Toksovo is 0.43, that of Razliv 4 and Razliv 6 is 0.49. Hepojarvi and Toksovo, Razliv 4 and Razliv 5 are sites evidently belonged to the same local group. Ornamental traditions of pottery of Okhta 1 and those of Hepojarvi and Toksovo demonstrate noteworthy resemblance, thus it is possible to conclude that these sites belonged to the same local group.

Keywords: Pit-Comb Ware; Hepojarvi; Toksovo; Okhta 1; Neolithic pottery ornamentation

1. Introduction

Neolithic sites of Hepojarvi (4480 ± 80 – 2020 ± 70 BCE Vereschagina 2003: 149) and Toksovo (dates unclear) are located in the area that is rather close to the source of Okhta river. In the estuary of Okhta river is located Neolithic site Okhta 1 (3660 – 2700 BCE Gusentsova, Sorokin 2011: 423). People who produced the Pit-Comb Ware in the considered region actively practiced fishing and used rivers and could easily travel from the area of Hepojarvi and Toksovo sites to the area of Okhta 1 site by the Okhta river, so it is possible to suppose that the sites of Hepojarvi, Toksovo and Okhta 1 could belong to the same local group. In order to estimate the degree of proximity Hepojarvi and Toksovo sites from one hand with Okhta 1 from the other hand is made a comparison of ornamental traditions of Okhta 1 with those of Hepojarvi and Toksovo.

2. The method of comparison

2.1. Theoretical background

The procedure of comparison of ornaments is basically the following: should be compared sets of elements that shape ornaments and positional distributions of elements of ornaments. However, such a method can be useful if we have entire vessels or can sufficiently restore entire ornamentations, but in many cases we have only potsherds which often can’t be arranged into an entire form. Having two randomly selected sets of randomly broken potsherds it is possible to conclude about the most frequent imprints only. The most frequent imprints are supposed to be the most characteristic imprints of a certain ceramic tradition. Thus, comparing the frequency of different imprints we can conclude about the proximity of ornaments of pottery and about the proximity of corresponding cultures. The method is about comparing randomly selected sets of potsherds which in their turn were randomly broken, so it can be named Monte Carlo method (see: Akulov, Nonno 2019).

2 CAES Vol. 5, № 2 (June 2019)

Fig. 1. Location of sites considered in the texts (lower map source: )

3 CAES Vol. 5, № 2 (June 2019)

2.2. An illustration of the procedure of comparison

Let’s imagine that we have the following sets of potsherds: A, B.

Set A:

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 /// /// /// /// /// /// /// /// /// /// /// /// /// /// /// /// /// /// ⁞ ⁞ ⁞ ⁞ ⁞ ⁞ ⁞ ⁞ ⁞ ⁞ ⁞ ⁞ ⁞ ⁞

The total number of potsherds is 20. There are two imprints: /// (1) and ⁞ (2).

Imprint 1: /// 18 elements: 1 – 18 Imprint 2: ⁞ 14 elements: 1 – 12, 19, 20.

It should be specially noted that one potsherd can belong to more than one class at once since one potsherd can bear different prints.

Percentage of imprint 1 is 18/20 = 0.9. Percentage of imprint 2 is 14/20 = 0.7.

Set B:

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 /// /// /// /// /// /// /// ⁞ /// ⁞ /// ⁞ ⁞ ⁞ ⁞ ⁞ ⁞ ⁞

The total number of potsherds is 15. There are two imprints: /// (1) and ⁞ (2).

Imprint 1: /// 9 elements: 1 – 9 Imprint 2: ⁞ 9 elements: 7 – 15

Percentage of imprint 1 is 9/15 = 0.6. Percentage of imprint 2 is the same 9/15 = 0.6.

Percentage of imprint 2 is 2/10 = 0.2; Percentage of imprint 3 is 5/10 = 0.5.

To estimate correlation degree of two sets we should do the following procedures: 1) to estimate correlation degree of sets of represented imprints, 2) to estimate correlation degree of percentages/distributions of common imprints, 3) take a superposition of two degrees of correlation.

The formula for correlation degree of two sets is the following:

4 CAES Vol. 5, № 2 (June 2019)

where: Nimp(A) – the number of imprints represented in A (first set), Nimp(B) – the number of imprints represented in B (second set), m – the number of common imprints.

A ~ B

The degree of correlation of imprints is 1. And thus the index of correlation of A and B is the following: 1*(0.6/0.9 + 0.6/0.7)/2 ≈ 0.76. (For more details see: Akulov, Nonno 2019).

3. Comparison ornaments from Hepojarvi, Toksovo and Okhta 1

3.1. Imprints of Hepojarvi

The set of potsherds of Hepojarvi is represented in fig. 2, fig. 3. The total number of potsherds is 26.

There are 8 imprints: 1) pit (samples 1, 2, 3, 4, 5, 6, 7, 9, 10, 11, 12 of fig. 2 and samples 17, 19, 21 of fig. 3). its percentage is 14/26 ≈ 0.54.

2) triangle pit (samples 13, 16, fig. 2) its percentage is 2/26 ≈ 0.07.

3) an imprint of a comb (samples 1, 2, 3, 6, 7, 9, 12 of fig. 2; samples 21, 23, 24, 25 of fig. 3), its percentage is 11/26 ≈ 0.42.

4) stroke (samples 4, 5, 8, 11, fig. 2 and sample 26, fig. 3), its percentage is 5/26 ≈ 0.19.

5) imprint of a rope (sample 15, fig. 2), its percentage is 1/26 ≈ 0.04.

6) triangle structure made of imprints of a comb (samples 17, 18, 19, 21, 22, fig. 3), its percentage is 5/26 ≈ 0.19.

7) circle (sample 20, fig. 3), its percentage is 1/26 ≈ 0.04.

8) imprint of a stick with teeth (sample 14, fig. 2), its percentage is 1/26 ≈ 0.04.

5 CAES Vol. 5, № 2 (June 2019)

Fig. 2. Potsherds from the site of Hepojarvi, part 1 (image source: Vereschagina 2003: 147)

Fig. 3. Potsherds from the site of Hepojarvi, part 2 (image source: Vereschagina 2003: 148)

6 CAES Vol. 5, № 2 (June 2019)

3.2. Imprints of Toksovo

Fig. 4. Potsherds from the site of Toksovo (image source: Gurina 1961: 439)

The set of potsherds of Toksovo is represented in fig. 4. 7 CAES Vol. 5, № 2 (June 2019)

The total number of potsherds is 18.

There are 3 imprints:

1) pit (samples 1, 2, 3, 4, 6, 7, 8, 12, 13, 14, 15, 16, 18, fig. 4), its percentage is 13/18 ≈ 0.72.

2) an imprint of a comb is represented upon 16 potsherds (all samples of fig. 4 except samples 1 and 15), its percentage is 16/18 ≈ 0.89.

3) stroke is represented upon 2 potsherds (1 and 15 of fig. 4), its percentage is 2/16 = 0.125.

3.3. Imprints of Okhta 1

Potsherds of the site Okhta 1 are represented in fig. 5.

The total number of potsherds is 12.

There are 5 imprints:

1) pit (samples 1, 2, 3, 5, 8, 11 fig. 5), its percentage is 6/12 = 0.5.

2) imprint of a comb (1, 6, 7, 8, 10, 12 fig. 5) its percentage is 6/12 = 0.5.

3) triangle pit (sample 4 fig. 5), its percentage is 1/12 ≈ 0.08

4) elongated pit (sample 6 fig. 5), its percentage is 1/12 ≈ 0.08

5) rhombic structures (sample 5 fig. 5) its percentage is 1/12 ≈ 0.08.

6) stroke (samples 2, 3 fig. 5) its percentage is 2/12 ≈ 0.16.

7) flower ornament (sample 9 fig. 5) its percentage is 1/12 ≈ 0.08.

3.4. Toksovo and Hepojarvi index of correlation

Index of correlation of ornaments of Toksovo and Hepojarvi is 0.43 (Akulov 2019: 29).

3.5. Hepojarvi and Okhta 1

There are 4 common imprints: pit, an imprint of a comb, triangle pit, stroke.

According to the formula represented in 2.2 index of correlation of Hepojarvi and Okhta 1 is the following:

(4/8 + 4/7)/2*(0.5/0.54 + 0.42/0.5 + 0.07/0.08 + 0.16/0.19)/4 = 0.54 * (0.92 + 0.84 + 0.87 + 0.84)/4 = 0.54 * 0.87 = 0.47 8 CAES Vol. 5, № 2 (June 2019)

Fig. 5. Potsherds from the site of Okhta 1 (image source: Gusentsova, Sorokin 2011: 441)

9 CAES Vol. 5, № 2 (June 2019)

3.6. Toksovo and Okhta 1

There are 3 common imprints: pit, imprint of a comb and stroke. Index of correlation of Toksovo and Okhta 1 is: (3/3 + 3/7)/2 * (0.5/0.72 + 0.5/0.89 + 0.125/0.16)/3 = 0.72 * (0.69 + 0.56 + 0.78)/3 ≈ 0.49.

3.7. Conclusion

The proximity of ornamental traditions of Okhta 1 and Hepojarvi, Okhta 1 and Toksovo are higher than that of Hepojarvi and Toksovo. The proximity of ornamental traditions of Okhta 1 and Toksovo is the same as that of Tarkhovka and Toksovo and the same as that of Razliv 4 and Razliv 5. The proximity of ornamental traditions of Okhta 1 and Hepojarvi is almost the same as that of Toksovo and Razliv 4. (Razliv 4, Razliv 5, Tarkhovka, Hepojarvi and Toksovo are supposed to be sites of the same local group/family.)

Fig. 6. Scheme showing degrees of proximity of sites of Sestroretskii Razliv (in current context different levels of proximity is marked by lines of different colors: red lines mark the highest degrees of proximity and blue lines mark the lowest degrees of proximity, for more details see Akulov 2019)

10 CAES Vol. 5, № 2 (June 2019)

R1 R2 R4 R5 R7 GR SG Tar Tok Hep R1 0.38 0.37 0.25 0.2 0.39 0.26 0.23 0.38 0.31 R2 0.52 0.28 0.31 0.35 0.37 0.67 0.41 0.37 R4 0.49 0.38 0.29 0.32 0.57 0.46 0.43 R5 0.5 0.25 0.25 0.41 0.22 0.27 R7 0.26 0.18 0.43 0.24 0.3 GR 0.14 0.34 0.41 0.24 SG 0.3 0.41 0.21 Tar 0.49 0.39 Tok 0.43

Table 1. Indexes of correlation of the following sites: R1 – Razliv 1, R2 – Razliv 2, R4 – Razliv 4, R5 – Razliv 5, R7 – Razliv 7, GR – Glinyanyi Ruchei, SG – Sosnovaya Gora, Tar – Tarkhovka, Tok – Toksovo, Hep – Hepojarvi.

Ornamental traditions of pottery of Okhta 1 and those of Hepojarvi and Toksovo demonstrate noteworthy resemblance, and thus it is possible to conclude that these sites of belonged to the same local group/family. It is possible to say that the sites of Hepojarvi and Toksovo were winter settlements while the site of Okhta 1 was the place where people from Hepojarvi and Toksovo practiced summer fishing.

References

Akulov A. 2019. Comparison ornaments of the Pit-Comb Ware from some Neolithic sites of the Southern part of Karelian Isthmus by Monte Carlo method. CAES, Vol. 5, N. 1; pp.: 2 - 32

Akulov A., Nonno T. 2019. Comparison ornaments of pottery by Monte Carlo method. CAES, Vol. 5, № 1; pp.: 33 – 38

Gurina N. N. 1961. Drevnyaya istoria Severo-Zapada evropeiskoi chasti SSSR (Ancient history of European part of USSR). Materialy i issledovaniya po arkheologii SSSR (Materials and researches on the archaeology of USSR), Vol. 87, Izdatel’stvo Akademii Nauk, Moscow – Leningrad

Gusentsova T. M, Sorokin P. Ye. 2011. Okhta 1 – pervyi pamyatnik epokhi neolita I rannego metalla v tsentral’noi chasti Peterburga (Okhta 1 – the first Neolithic – Early Metal Period site in the central part of St. Petersburg). Rossiiskii arkheologicheskii ezhegodnik (N 1, 2011); pp.: 421 – 451

Karelian Isthmus https://en.wikipedia.org/wiki/Karelian_Isthmus – accessed June 2019

Vereschagina I. V. 2003. Poseleniye Khepoyarvi v yuzhnoi chaste Karel’skogo peresheyka (The settlement of Hepojarvi in the southern part of Karelian isthmus), in Timofeev V. I., Sinitsyna G. V. (eds.) Neolit – eneolit yuga i neolit severa Vostochnoi Evropy (Neolithic – Eneolithic periods of the South of Eastern Europe and Neolithic period of Northern part of Eastern Europe). Institute of Material Culture History Publishing, Saint Petersburg; pp.: 140 – 151