7.5.1. Inorganic Materials

Ceramic Production

No native wheelmade pottery is known from the Iron Age, though the possibility that some may have been finished on a slow wheel has been postulated (ScARF Iron Age section 4.5). There are also no obvious kiln sites in the Highlands, though kilns are only really needed for large-scale pot production; domestic pots could be fired in a version of a controlled bonfire. Work elsewhere suggests local clay was the norm (ScARF Iron Age section 4.4), and this is likely for the Highlands as well.  The source of the clay for the Late Bronze Age to Early Iron Age clay moulds used for metalworking at Bellfield, North Kessock was investigated, and a local source found most likely. Interestingly the analysis showed different clay sources for moulds and for daub (Clark et al 2017, 8–10). Although there were relatively few ceramics found at Clachtoll broch (MHG13002), most of the surviving material shows an affinity with Hebridean style pottery, and the evidence suggests the pottery was being manufactured locally (Graeme Cavers pers comm).

Metalworking: Iron

The evidence for iron working in Iron Age Scotland was reviewed by Cruickshanks (2017) which included key sites of High Pasture Cave (MHG32043), Culduthel (MHG49950), Seafield West (MHG3058), Applecross (MHG7680) and Clachtoll Broch (MHG13002). In earlier excavations slag was not always retained or reported; nevertheless, there is widespread evidence for ironworking surviving in the Highlands (see Table 7.11). Recent excavations over the past decades in the Highlands have provided well-dated evidence of ironworking occurring in a variety of contexts (Hunter 2015, 233). The timing and means by which this new technology reached the Highlands still needs to be examined (Hunter 2015, 238). A key issue is differentiating iron smelting (the making of iron) from smithing (its subsequent forming). This can be done with expert visual analysis backed up where necessary with scientific work.

High Pasture Cave (MHG32043; Case Study High Pasture Cave) provides evidence of some of the earliest iron smelting in Scotland and Britain as a whole, with a furnace fragment dating from the 8th–5th centuries cal BC; the site also provided some of the earliest iron artefacts surviving in Scotland. The centuries of deposition and movement of material make it difficult to know if the remains were found where activities actually took place, as it appears that some of this waste may have been deliberately deposited (Cruickshanks 2017; Birch et al forthcoming).

At Culduthel there was extensive ironworking, both smelting and smithing. Five iron furnaces were found within the suggestions of a baked wattle and daub superstructure. The site has produced one of the largest Iron Age iron assemblages in Scotland. It has evidence of the entire iron-working process, from smelting iron ore to smithing. Some of the furnaces are well preserved, a rare find for Iron Age Scotland, showing free-standing, non-tapped furnaces. Fragments of the fired clay chimney with wattle impressions were uncovered; this is unique in Iron Age Scotland. Three different types of tuyères for the bellows were found. Scientific analysis of the slag and the iron shows that these furnaces produced carbon steel. Iron working on the scale shown at Culduthel would have required substantial charcoal as fuel, suggesting that the surrounding woodland may have been managed to support this. Wood used varied and included alder, hazel, oak and wild cherry. The ‘Fire in the Hole’ UHI project is currently looking at the issues of wood use and woodland management for the Culduthel industries (Scott Timpany pers comm). Culduthel’s importance in understanding Iron Age ironworking cannot be underestimated (Hatherley and Murray 2021).

archaeologist in a high vis vest excavating round stone structures
Excavating the iron smelting furnaces at Culduthel. ©Headland Archaeology (UK) Ltd

Bog iron is plentiful in much of the Highlands, and was the key resource used in ironworking. Metalworking required fuel which could create and sustain high temperatures. Wood charcoal was available in varying degrees throughout the Highlands, but peat could also be used as it was plentiful as well. At the smithing hearth at Seafield West, oak was predominant and it was probably preferentially chosen for this activity (Cressey and Anderson 2011, 32–3).

Metalworking: Non-ferrous

The evidence for non-ferrous metalworking was reviewed by Heald (2005); since then a number of excavated sites have added to this picture. The presence of non-ferrous metalworking is generally taken to indicate a high-status settlement that is able to obtain raw materials, but that also had technical expertise, perhaps from itinerant specialists. Recent analysis of the metal in the Poolewe hoard which dates to the very Early Iron Age showed that new sources of metal were coming into the area at the beginning of the period (Knight et al 2021; Case Study Poolewe Hoard).

It is unclear if Scottish copper sources were used in the Highlands. Tin would have been imported or obtained from recycling. Zinc is found in copper alloys from the Roman Iron Age, deriving presumably from remelting Roman objects (ScARF Iron Age sections 4.4). Copper alloy ingots preserving marks of hammering, were found at Carn Liath broch, Golspie (MHG10872). More work on changing alloys, with dating and geographical use, is needed.

Ingot of zinc bronze from Carn Liath at Golspie, Sutherland. It preserves marks from hammering, made sometime between 100 and 400 AD. ©National Museums Scotland

The evidence from Bellfield, North Kessock shows that similar types of copper alloy objects to those found in the Late Bronze Age continued to be manufactured in the earliest Iron Age (Knight et al 2021; Clark et al 2017). At Culduthel, there was evidence for copper alloy working, as well as iron working. This included tools and moulds for decorative objects, as well as the largest recorded assemblage of crucible and mould fragments in Iron Age northern Britain. These include an ingot mould made from a re-used rotary quern (Hatherley and Murray 2021).

Hunter (2015, 234–5) has argued that the evidence in Scotland shows that there were regional differences in craft organisation, with non-ferrous metalworking being relatively common in the Highlands compared to other areas further south. This suggests that production was not tightly controlled in the north. There is also evidence of sharing styles and techniques over large areas.

Site Area Dating Comments Source
Crosskirk broch C Site with long period of use Slag and two crucible fragments MHG39521; Fairhurst 1984
Keiss Road broch C   19th century excavations. Moulds for non-ferrous metalworking, but dating uncertain MHG1650; Heald and Jackson 2001
Keiss Harbour broch C One comb dated to early medieval period 19th century excavations. Non-ferrous metalworking: crucible for melting tin-bronze with inclusions of lead and zinc; dating uncertain MHG1659; Heald and Jackson 2001
Nybster broch C 1st to 3rd centuries AD Non-ferrous metalworking (crucibles; possible pin mould). No evidence iron metalworking MHG1593
Skirza Head broch C   19th century excavations included one lump of slag MHG655
Langwell S   Vitrified dun, with slag in pit for Stage 5 occupation, perhaps early years AD, and more slag undated outside the dun MHG7371; Nisbet 1994-1995
Carn Liath broch S   Ironworking evidence (slag) from early investigations MHG45699
Dun Morangie, Tarlogie ER Early centuries AD Ironworking (slag). Awaiting final report MHG44719; Hatherley 2015b
Dalmore ER Pits with hammerscale, at site with roundhouses and grain storage pits. MHG17924
Bellfield, North Kessock ER Awaiting publication. 4th to 2nd centuries BC Ironworking, furnace linings and slag; awaiting final report for datings as the site is multi-period MHG53532; Murray 2012; Hatherley & Scholma-Mason forthcoming; Dates reported in Hatherley and Murray 2021
Seafield West I 180 BC – AD 70 Ironworking debris and hearth linings; slag and hammerscale MHG3058; Cressey and Anderson 2011; GU-8032
Culduthel I Various. Extensive ferrous and non-ferrous metalworking including furnaces MHG49950; Hatherly and Murray 2021; Case Study Culduthel Iron Age Settlement
Comar Wood Dun I Nine dates, one group 357–184 BC, the other first centuries AD A few pieces of slag suggest limited metalworking MHG55867; Peteranna and Birch 2017a
Dun Lagaidh WR   Copper alloy sheet debris MHG45426; MacKie 2007
Applecross broch WR 4th to 3rd centuries BC to early centuries AD Smithing hearth with tuyere fragment, quantity of slag; further report awaited MHG7680; McCullagh 2012
Clachtoll broch NWS 50 BC to early years AD Possible bloom cake. No hammerscale. Probably not taking place in the broch MHG13002; Graeme Cavers pers comm; Case study Clachtoll broch
High Pasture Cave Skye Good dating Range of metalworking; Awaiting publication. MHG32043; Birch et al forthcoming; Case Study; High Pasture Cave
Uamh an Eich Bhric, Fiscavaig Skye A number of dates showing occupation c AD 50–150 Slag; casting debris for non-ferrous metalworking, fragments of furnace lining MHG51768; Wildgoose and Birch 2009
Creag A’Chapuill Skye   Cave excavated in 1930s with multi-period finds including Iron Age pottery and slag. Good preservation MHG4898
Coille Gaireallach Skye AD 80–240 Slag in hearth; date from charcoal also in pit MHG59525; Wildgoose 2016, site LS11; SUERC-33648
Dun Flodigarry Skye 178 BC–AD 132 Slag; context and dating uncertain MHG5219; GU-1662
Dun Artreck Skye Multi-period Slag; crucibles for non-ferrous metalworking MHG5019; MacKie 2000
Dun Beag Skye   Finds typical of brochs, and include iron slag and a crucible, but also medieval finds, so dating not secure. MHG5018
Dun Telve broch Lochalsh   1914 excavations: Two fragments of iron slag MHG5355
Dun Deardail hillfort L 5th to 2nd centuries  BC Slag, hammerscale; crucible for non-ferrous metalworking MHG4348; Ritchie 2018; Case Study; Dun Deardail Hillfort
Table 7.11  Highland sites with Iron Age metalworking evidence

Glass Making

While certain types of glass beads are diagnostic objects for the Iron Age period (see 7.4), evidence for their manufacture is rare, although there are regional styles for both glass beads and enamel production (Hunter 2015, 235). Production at Culbin Sands has been proposed based on finds of glass slag (Guido 1978, 85–7). The evidence from Culduthel is very important, as it provides the first secure evidence for glass working in Scotland: opaque red for metalwork inlay and yellow, blue and clear glass for jewellery, mainly beads. There, glass ingots of imported Mediterranean glass were used. The main products were Guido Class 8 and Class 13 beads. This activity is dated to some point between the 2nd century cal BC and early 1st century cal AD. Analysis of the debris has provided a number of insights into production (Davis and Freestone 2018; Hatherley and Murray 2021; Hunter 2021).

Hunter (2021) argues that glass working may have occurred more widely than the current evidence suggests, as it required little space, and the evidence is small droplets that are easily missed in the field. He also emphasises the links with non-ferrous metalworking. The similarities of beads and other objects from Culbin Sands and Culduthel suggests the possibility that glass workers, and possibly bronze workers, were specialists that moved through the area (Hatherley and Murray 2021).

Stone Working

Some areas of the Highlands have stone resources which were exploited in the Iron Age. The concentration of objects made from steatite (Map 7.5), an easily carved stone with good heat retentive properties, suggests local extraction and production, although more work needs to be done identifying local outcrops (Hunter 2015, 230). Steatite in Scotland is generally considered to be from Shetland (ScARF Iron Age section 4.4), but scientific analysis can compare objects with outcrops, and could therefore determine if this is the case for the Highland objects. Talc was also used in the Highlands as a temper for pottery; at Lairg the source was probably around 30 km away, suggesting local knowledge and organisation (MacSween and Dixon 1998). The Highland examples should be studied, with attention towards any which seem unfinished and that might indicate possible production sites. Non-destructive XRD analysis maybe able to show correlation between production sites and finished objects.

Jet-like materials such as cannel coal, lignite and albertite are also available in the Highlands, with production evidence from Carn Liath, Golspie (MHG10872) and Upper Suisgill (MHG9345; Sheridan et al 1998); the Golspie oil shale source appears to have been the key one for the Iron Age Highlands. There appears to have been movement of the raw materials as well as finished objects. Different communities appear to be and working Brora shale and removing cores for bangles in at least three different ways. This suggests widespread production rather than controlled access (Hunter 2015, 231). There is no evidence in Scotland of the use of a lathe for bangle production as in southern England (ScARF Iron Age section 4.5).

Roughouts to make shale bangles from Carn Liath. ©National Museums Scotland

Querns may have been produced locally, but the known Highland quern quarries at Rubha nam Brathairean, Skye (MHG52601) and possibly Brunachan, Lochaber (MHG4245) are undated.

Painting Painted Pebbles

Painted pebbles which are small smooth, painted quartz stones, are a diagnostic object for the Iron Age (see 7.4). Scientific analysis combined with experimental attempts to recreate the pigmentation suggests the use of either haematite or lamp black. Lamp black may have perhaps been made from soot collected from burning organic material of vegetable composition such as peat. Feathers and stems of plants were both found to produce markings similar to those on the pebbles. This work shows that the soot from fires could well have been used not only for decorating pebbles, but also for other uses where the products no longer survive (Arthur et al 2014; Carver et al 2016, 262ff).



Case Study: Poolewe Hoard

Comments 1

  1. Post-excavation at Clachtoll broch has investigated some sources of steatite in Sutherland, to see if the material of any of the many vessels found at the site could be sourced. The results show the objects were not from Shetland steatite, but also not from any of the sites sampled. Described by Andy Heald in the Clachtoll broch conference in October 2021 ( https://vimeo.com/625560343/6e65d9cdfd ), with full details to be included in the final report.

Leave a Reply