Journal of Human Genetics (2012) 57, 467–469

& 2012 The Japan Society of Human Genetics All rights reserved 1434-5161/12 $32.00
www.nature.com/jhg

CORRESPONDENCE

‘Early Neolithic’ graves of the Carpathian Basin

In the Journal of Human Genetics, volume 56, and, among others, a cemetery of the Sar- the excavation report it was stated that graves
issue 11(15 November 2011), an article about matian period (1st–4th centuries AD) as well 2 and 3 belonged to the Early Neolithic Körös
the Early Neolithic genetic composition of as graves of the Hungarian Conquest period culture.5 About grave 8 the following
the Carpathian Basin was published. The (10th century AD). The skeletons involved in description can be found in the 1976
article claims that some of the individuals the Guba et al1 article come from two graves, excavation documentation (pp. 2–3): ‘Each
examined in the course of this study belong Nos. 20 and 22, both indicated as Early of the graves 5–8 were disturbed, none of
to the mitochondrial haplogroup (hg) N9a, Neolithic. In fact grave 22 is Sarmatian with them belonged to the Körös culture. Grave
which is common in East Asian and virtually special Roman pottery, (terra sigillata) placed No. 8: a skeleton disturbed and robbed, the
absent in European/West Eurasian popula- near the skeleton as a grave good.2–4 bones are not in anatomic order. Grave
tions. The inferred discontinuity between the According to the excavator, I. Juhász, grave goods: a piece of bronze, a spiral of a
first farmers in Hungary and the modern 20 is in fact a Hungarian Conquest period bronze fibula, several glass and chalk pearls.
population is specifically based on the (Magyar) skeleton buried with a horse, dated The grave can be dated to the Sarmatian
appearance of this haplogroup, hitherto to the 10th century AD. period’.
absent in the European Neolithic dataset. (II) Csongrád–Bokros, with one grave (IV) The individual from Kisköre–Gát can
Here we wish to point out a number of claimed to belong to the Neolithic ‘ALP’ probably be assigned, and one specimen from
serious mistakes in the dating of the skeletons (Alföld Linear Pottery end of 6th Mill. BC) Szegvár can definitely be assigned, to the Late
investigated as well as the phylogenetic clas- culture. However, this site is also a Sarmatian Neolithic Tisza culture (5000–4500 BC)
sification of the generated mitochondrial (1st–4th centuries AD) settlement. According instead of the indicated earlier periods of
HVS-I sequences, which have a significant to the documentation of the excavation (7–8 the Neolithic.6,7 The Late Neolithic dating of
influence on the interpretation of the genetic August, 1979, pp 35–37) the following is the Kisköre–Gát grave is estimated by the
data. reported: ‘Inside Pit No. 20, at a depth of - excavator, on the basis of its position.8
A number of well-known Neolithic sites 50 cm ... parts of a human skeleton were Consequently, six out of the eleven graves
are involved in this study, such as Szarvas 23, found. The remains consisted of a left arm considered to be Neolithic were clearly mis-
Szakmár–Kisülés and Csongrád–Bokros. and left leg, as well as some ribs... the fill of dated. In addition, six individuals of this
However, all these sites were occupied in the pit contained both Neolithic and Sarma- dataset were attributed to an East Asian hg
several archaeological periods, prehistoric tian pottery fragments, Sarmatian sherds (three N9a, one D1/G1a1, one M/R24 and
and early historic. With a multi-period site, were found even below the skeleton’. An oral one C5). Obviously, the presence of East
the skeletons have to be handled with special communication was added by the archaeol- Asian hg in European sites is unusal and
care to the archaeological contexts. If there is ogist LA Horváth, who studied the findings presents an important finding for the recon-
any doubt about the date, it is essential to of the site: ‘the fragmented skeleton was struction and timing of past population
establish the absolute date of each skeleton considered Sarmatian already during the movements. However, exactly the three indi-
investigated. In the article in question, the excavation’. viduals with the N9a hg (Szarvas 23 grave 20,
following examples are wrongly dated: (III) Szakmár–Kisülés, with one grave Szakmár–Kisülés 8, and Csongrád–Bokros,
(I) Szarvas 23, with two graves claimed to claimed to belong to the Neolithic ‘KSC’ grave inside Pit 20) can be considered to
belong to the Early Neolithic ‘KSC’ (Körös– (Körös–Starčevo–Cris¸) period. This site is a the Sarmatian and, in one case, early Hun-
Starčevo–Cris¸) period. Szarvas 23 is a multi- settlement of the Körös culture, which was garian (Magyar) period. Thus, all N9a speci-
period site, with findings from the Early disturbed by diggings of a series of later mens from the Early Neolithic dataset (the
Neolithic Körös culture (6000–5450 cal BC) periods, including a Sarmatian cemetery. In first half of the 6th millennium BC)
 Correspondence
468

Table 1. Summary of published and revised archaeological and genetic information

Specimen Dating Revised dating Haplotype hg Revised hga (East) Asian hg

Szarvas 23/22 KSC Sarmatian 16235G 16261T 16291T 16293G 16304C Novel H?
Szarvas 23/20 KSC Hung. Conq. p. 16223T 16257A 16261T N9a N9a þ
Szakmár–Kisülés8 KSC Sarmatian 16223T 16257A 16261T N9a N9a þ
Csongrád–Bokros 20 ALP Sarmatian 16223T 16257A 16261T N9a N9a þ
Ecsegfalva 23A ALP ALP 16147a 16172C 16189C 16223T 16248T 16274A 16355T N1a N1a1
Kisköre–Gát 15 ALP Tisza? 16223T 16325C 16362C D1/G1a1 D1/G1a1 þ
Vörs 52 KSC Starčevo 16223T 16288C 16298C 16327T C5 C5 þ
Mezo+ kövesd 25 ALP ALP 16324C M/R24 H? þ
Aszód Late Neolithic Tisza — H H/HV/U/R
Folyas ALP ALP — H H/HV/U/R
Szegvár KSC Tisza — H H/HV/U/R

Abbreviations: ALP, Alföld Linear Pottery culture; Hung. Conq. p, Hungarian Conquest period; hg, haplogroup; KSC, Körös–Cris¸–Starčevo culture.
Misdating and insufficient haplogroup classification of the samples are shaded in grey. The last column summarise the influence of the revison on (East) Asian haplogroups in the Neolithic dataset.
awww.phylotree.com, built 14 (5 Apr 2012).

originated from a much younger period than In contrast to the misdating itself, we do The problems discussed represent a caveat
reported: the 1st millennium AD. Addition- not question the integrity of the finding of hg for all contributors to the archaeogenetic
ally the D1/G1a1 individual from Kisköre– N9a in a Sarmatian and Magyar context. literature. Mistakes can occur when authors
Gát was also misdated. Sarmatians migrated in the early first Millen- of archaeogenetic studies neglect both the
Furthermore, we have detected inaccura- nium AD, Magyars in the late first Millen- archaeological context of the skeletons stu-
cies concerning the hg classification of the ium, from the steppe region of present Russia died and the archaeological literature about
mitochondrial sequences. In three graves, and Ukraine to Eastern Hungary. The highest the sites involved. While one cannot expect
the HVS-I sequences are insufficient for frequency of the published N9a haplotype molecular biologists to scrutinise minor
a definite classification into the known mito- (16223T 16257a 16261T) can be found in excavation reports, let alone hunt for unpub-
chondrial phylogeny (Szarvas 23 grave modern populations of East Asia (Vietnam lished documentation, it is all the more
22, Kisköre–Gát 15 and Mez+okövesd 25). (0,78%), China (0,23%), Thailand (0,14%)) important to integrate archaeologists into
In these cases, the authors listed all possible as well as Central Asia (Kyrgyzstan (0,28%), the research process. This would not only
hg or indicated the sequence as ‘novel’. Kazakhstan (0,22%), Uzbekistan (0,21%) solve the problem of dating a skeleton 6000
However, in some cases these assignments and the Buryat Republic (0,13%)) (references years earlier than its real age, but would also
are incorrect. The haplotype of Szarvas 23 available upon request). Thus, the observa- share responsibility for the results and estab-
grave 22 should not be determined as ‘novel’ tion of high N9a frequencies in a Sarmatian lish real interdisciplinarity between archaeol-
because this sequence most likely belongs to and Magyar population may well support an ogy and the natural sciences.
the European hg H.9 Similarly, the 16324C (East?) Asian affinity.
haplotype of the Mez+okövesd individual was The data presented in this study were ACKNOWLEDGEMENTS
The authors wish to thank I Juhász and
previously described from several studies as claimed to be the first archaeogenetic dataset
LA Horváth for their kind oral information about
hg H.10 As a result, these unspecific of Neolithic cultures of the Carpathian Basin.
the Szarvas grave 20 and Csongrád grave in
haplogroup classifications introduced a However, the fact that previously published Pit No. 20 datings.
higher frequency of East Asian hg. In all data were included is not obvious in the
these cases, additional analysis of diagnostic article, which is misleading. Some of these Eszter Bánffy1, Guido Brandt2 and
coding region SNPs are required in order to samples have already been published (Ecseg- Kurt W Alt2
unambiguously assign the correct hg. This falva 23A11 and Szarvas 23 grave 2012). The
1Institute
of Archaeology,
also includes the three CRS individuals latter article by Burger et al mentions the
reported as hg H (Aszód, Folyás 111 and very same grave (Szarvas 23 grave 20), Research Centre for the Humanities,
Szegvár 25), which also could belong to the presented in the paper by Guba et al as Hungarian Academy of Sciences, Budapest,
hg HV, R or U. belonging to the Early Neolithic Körös Hungary and 2Institute of Anthropology,
Owing to the fact that six skeletons of the culture, while they are, as shown above, Johannes Gutenberg University Mainz,
supposed Neolithic dataset had been mis- 6000 years later. In the Burger et al study Mainz, Germany
dated, and some haplotypes had been insuf- this misdating occurs repeatedly, with yet E-mail: [email protected]
ficiently classified, we cannot support the another, Sarmatian grave (Szarvas 23, grave 4).
conclusion of a high frequency of East Asian All these cases of misdating could have
hg (especially N9a) in Neolithic cultures of been avoided by applying 14C dating. In 1 Guba, Z., Hadadi, É., Major, Á., Furka, T., Juhász, E.,
Koós, J. et al. HVS-I polymorphism screening of
the Carpathian Basin. In fact, only two the article by Guba et al. there is a general ancient human mitochondrial DNA provides evidence
Neolithic individuals remain with an East reference to 14C dates,13 but this study is for N9a discontinuity and East Asian haplogroups in
the Neolithic Hungary. J. Hum. Genet 56, 784–796
Asian hg (Vörs, hg C5 and Kisköre–Gát 15, focused on another period. Unfortunately, all (2011).
hg D1/G1a1) (Table 1) while only Vörs can efforts to produce authentic ancient DNA 2 Jankovich, B. D., Makkay, J. & Szo+ ke, B. M.
be ascribed to the Early Neolithic (Starčevo data are worthless if the archaeological (eds) Magyarország régészeti topográfiája 8. A szarvasi
járás (The Archaeological Topography of Hungary 8. -
culture). In the light of all these facts both background to a genetic project is missing the Szarvas district) (Akadémiai Kiadó, Budapest,
results now seem questionable. or insufficiently explored. 1989).

Journal of Human Genetics

3 Juhász, I. Mázas római edény Szarvasról. A Békés 7 Korek, J Szegvár-Tu+zköves, in The Late Neolithic of the PLoS Genet. 3, e104 (2007) Erratum in: PLoS Genet.
megyei múzeumok közleményei 23, 89–97 (2002). Tisza region (ed L, Tálas) 47–60 (Kossuth Press, September 14;3(9):1785 (2007).
4 Makkay, J. The excavations of the early neolithic sites Budapest-Szolnok, 1987). 11 Haak, W., Forster, P., Bramanti, B., Matsumura, S.,
of the Körös culture in the Körös valley, Hungary: the 8 Siklósi, Zs. A társadalmi egyenlo+ tlenség nyomai a késo+ Brandt, G., Tänzer, M. et al. Ancient DNA from the first
final report. Vol. I: The excavations: stratigraphy, struc- neolitikumban a Kárpát-medence keleti felén (Traces of European farmers in 7500-year-old Neolithic sites.
tures and graves (Societa per la preistoria e protoistoria social inequality in the Late Neolithic, in the eastern Science 310, 1016–1018 (2005).
della regione Friuli-Venezia Giulia, Trieste, 2007). half of the Carpathian Basin). Unpublished PhD dis- 12 Burger, J., Kirchner, M., Bramanti, B., Haak, W. &
5 Bognár-Kutzián, I. Ausgrabungen in Szakmár-Kisülés sertation, ELTE Institute of Archaeological Sciences, Thomas, M. G. Absence of the lactase-persistence-
im Jahre 1975. Mitteilungen des Archäologischen Budapest, (2010). associated allele in early Neolithic Europeans.
Instituts der Ungarischen Akademie der Wissenschaf- 9 Sykes, B. Blood of the Isles (Bantam Press, London, Proc. Natl Acad. Sci. USA 104, 3736–3741 (2007).
ten 7, 13–17 (1977). 2006). 13 Hertelendi, E., Svingor, É., Raczky, P., Horváth, F.,
6 Korek, J Die Theiss-Kultur in der mittleren und nördli- 10 Behar, D. M., Rosset, S., Blue-Smith, J., Balanovsky, Futó, I. & Bartosiewicz, L Durationof tell settlements
chen Theissgegend (Inventaria Praehistorica Hungariae O., Tzur, S., Comas, D. et al. The Genographic Project at four prehistoric sites in Hungary. Radiocarbon 40,
III, Budapest, 1989). public participation mitochondrial DNA database. 659–667 (1998).

Journal of Human Genetics