The author of this, the most recent of the York Archaeological Trust's fascicule series, sets himself a substantial challenge — to present evidence from 56 sites spread around the environs of the Roman fortress and colonia. The fact that they were excavated any time between 1976 and 2005 and range from small, salvage work only a few metres across, to major projects such as Coppergate, investigated systematically over several years, only adds to the task, as does the uneven distribution of the work: a product of the vagaries of modern development. Yet tackling such complexity will be a growing challenge for urban archaeologists: PPG16's emphasis on in-situ preservation means that our evidence will increasingly derive from small holes dug into the Roman deposits, sometimes stopping just at the point where the archaeology gets interesting.
In order to give the process some sort of order, Ottaway divides the area around York's urban core into seven zones. Detailed descriptions are presented for sites in each, and their collective implications then summed up in terms of roads, landuse/settlement and burials (sequence diagrams summarising each site might have helped move from detailed evidence to zonal synthesis). A final chapter draws threads together across zones in a chronologically ordered discussion. In a publication of this complexity, some small inaccuracies inevitably slip through: the final Zone 1 site is Water Lane Clifton, not CCTV pit Bootham; trench detail at 45–57 Gillygate does not match the site location plan; 16–22 Coppergate is not numbered on fig. 128; Trench 10 at Terry's factory cannot be located. But, generally, his system works.
A picture emerges of limited development of Eboracum's environs for perhaps 100 years after the conquest: little beyond second-century ceramic production near the fortress and early roads set out along high ground above glacial moraine (understanding York's development would benefit hugely from a proper grasp of detailed drift geology). Observations of thoroughfares sometimes modify earlier (1962) Royal Commission interpretations (the road entering the city from the south-west lay 30 m west of the medieval and modern access point, not beneath it) and sometimes refute them (their conjectural ‘Road 2’ running south of the fortress does not exist, at least on the line published there).
Later in the second century, glass-making at Coppergate and Ebor Ware pottery produced elsewhere mark an economic upturn, linked by O. with the return of the legion from Hadrian's Wall. These decades also evidence an increased number of subsidiary streets, plus boundary ditches defining adjacent landholding. Within the latter, newly defined areas, rubbish-dumping is intermingled initially with burials, and it is only in the third century that such activities are controlled and cemeteries formalised: monumental burials such as the mausoleum at Blossom Street; more coffined-inhumations, sometimes with added gypsum; and the well-known decapitations inserted to the south-west, clearly successive insertions rather than a product of a single event.
This success story is not carried through, however. Some of these cemeteries, rather than developing in linear fashion away from the town, lie closer to it than their forerunners — expectations for York's development unfulfilled? Sometime after the end of the third century, road metallings suggest a narrowing of main thoroughfares, while the general absence of diagnostically late fourth-century pottery such a Crambeck Ware from sites beyond its core suggests that York was becoming a very different, perhaps much shrunken, place well before the formal ‘end of Roman rule’.
With such complex evidence, other interpretations are naturally possible: cobble spreads in fig. 111, for instance, seem to obey the positions of ditch, ‘mortary surfaces’ and inhumations, not to belong to a different phase, while the anomalously aligned walls at Coppergate (fig. 138) could comprise boundaries defining burial areas contemporary with the inhumations to the north, not a ‘building’ of peculiar plan form. But at least the evidence presented here allows such alternatives to be postulated.
Yet publication of this form raises other issues. Splitting fascicules by evidence type speeds up dissemination but divides stratigraphy and space from dating and function (fortunately, here, Monaghan's Roman pottery was published in 1993), while separating Roman and Anglian periods allows discussion of sub-Roman transitions to slip through the net. The considerable evidential detail here is also inconsistent in its coverage: some trenches are not even discussed, other deposits are not described — paradoxically, because large sites like Coppergate would make presentation too cumbersome. Thus small trenches are given in detail but potentially more coherent evidence is left in archive. The Trust has a coherent system for storing all excavation records (IADB) and a website for accessing detailed site evidence (http://www.yorkarchaeology.co.uk/resources/ayw.htm). Perhaps the time has come to have faith in these as sources of underlying evidence and allow formal publication to concentrate purely on interpretation and synthesis.
