Agria 39. (Az Egri Múzeum Évkönyve - Annales Musei Agriensis, 2003)
Domboróczki László: Radiokarbon adatok Heves megye újkőkori régészeti lelőhelyeiről
So it was that in the middle of the 1990s, studies started using absolute means of dating with ever greater regularity. Thus far, these early studies have not really been evaluated or criticised, and it is for this reason that we would like to devote some time to this question now. 51 In their study published in 1994, F. Horváth and E. Hertelendi used mean С 14 datings for their early and mid-Neolithic chronological clusters in eastern Hungary. The Körös Culture sites occupied the first three clusters, and the ALP sites clusters IV-VIII. Using the clusters they had drawn up, the ALP was divided up into four parts from 5400 to 5000 ВС, each lasting 100 years (V and VI having been brought together), before each of these periods was paired off with those phases existing in the surrounding cultures. In doing so the authors tried to bring N. Kalicz and J. Makkay's chronological scheme, consequently much adapted by themselves and P. Raczky, into line with the С 14 dates. In the process they brought attention to further contradictions in the sequence of the regional groups and the issue of regional continuity. 52 Partly because they didn't have any data for the early Szatmár II sites, Horváth and Hertelendi placed the Szatmár II group (or to put it otherwise, the first phase of the ALP) after the earliest phase of TLP (Transdanubian Linear Pottery) and after the Körös Culture's final protovinca phase. 53 As they only used mean values from the С 14 data in their chronological system, without the benefit of standard margins of error, their method came in for some justified criticism. 54 The margin of error, which occasionally exceeded ± 100 years, meant that within the sigma 1 interval there was scope for variations of at least 200 years, making the existence of a 100-year cluster seem somewhat contrived. As the Gubakut data has already proved, the use of mean values does not, however, automatically mean an error will be made. Although remote, the possibility that they are accurate nevertheless remains. Another point, and one not emphasised sufficiently in the critical remarks, is that a site cannot be dated on the basis of one single piece of data. As we saw at Gubakút, some sites were inhabited continuously for as long as three hundred years. It is for this reason, that the validity of F. Horváth and E. Hertelendi's clusters should be called into question even more. By taking single samples, however, one can only say that the site was inhabited at that particular time. Not much more can be stated about the lifespan of the entire settlement. This is something one would be wise to consider in what follows. Perhaps the best way of treating С 14 data is to insist that we always quote the identification number of the objects we are talking about. When dating individual objects the standard deviation is now already lower, because of the ±60 year differential, making it possible to fit an object into a one-hundred-year cluster. But in the case of larger sites the various objects will probably fall into different clusters. The approach for us in our particular region would therefore be to treat each object as a datable unit, and then having done that house, or at the very least, pit chronologies will emerge. Around the middle of the 1990s, as a result of the rapid growth in the data available, further evaluations took place, mainly concerning the tell cultures of the late Neolithic 51 With the exception of P. Raczky's opinions expressed during F. Horváth's defence of his doctoral thesis. 52 HORVÁTH Ferenc-HERTELENDI Ede 1994. 114-118. 53 F. Horváth and E. Hertelendi, like J. Pavúk (PAVÚK, Juraj 1994. 139) suspected that the development of the TLP preceeded that of the ALP. (HORVÁTH Ferenc-HERTELENDI Ede 1994. 118). For further literature offering an alternative opinion: DOMBORÓCZI 1997c. 73., footnote 55. 54 P. Raczky drew attention to this during the course of F. Horváth's doctoral defence. 22
