Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 172 Introduction Dingwall is the name of a modern town in Highland Scotland, but is first recorded in A.D. 1226 (Dingwell) and has for over a century been identified with ON þing-völlr, “field of the assembly” (Fig. 1; Fellows-Jensen 1993:55–56, 1996:24; Johnston Re-evaluating the Scottish Thing: Exploring A Late Norse Period and Medieval Assembly mound at Dingwall Oliver J.T. O’Grady1,*, David MacDonald†, and Sandra MacDonald2 Abstract - In this paper, we make a case for identification of a thing site at Dingwall in Scotland, which has previously only been known from place-name evidence. A complex of features associated with the thing is reconstructed through reference to a mound, known in the medieval period as the Mute hill of Dingwall, which is shown to have been closely associated with a legally bounded field and church. Results are presented from a detailed local historic landscape study with findings from the first modern archaeological exploration at the candidate assembly mound, including geophysical survey and excavation. We discuss the complexities of the site’s historical and landscape context with reference to the expansion of Norse lordship into northern Scotland during the Viking Age and Late Norse Period, and a review of recently identified thing sites elsewhere in Scotland. A considered interpretation is achieved of the political context for the thing’s establishment and reuse during the medieval period, with reference to radiocarbon dates from the mound and discussion of the potential for Late Norse and Early Gaelic influences. Debating the Thing in the North: The Assembly Project II Journal of the North Atlantic 1Almond Cottage, Glenalmond, nr Perth, Perth and Kinross, UK, PH1 3RX. †Deceased. 24 Castle Gardens, Dingwall, Highland, UK, IV15 9HY. *Corresponding author - Oliver.O’Grady@glasgow.ac.uk. 2016 Special Volume 8:172–209 Figure 1. Map of Dingwall’s location in the northern world. (Background map reproduced from Ordnance Survey map data by permission of Ordnance Survey. © Crown Copyright 2016.) Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 173 1934:156; Southey 1929:118; Watson 1904:93). The name is comparable with a small number of other “ding” place-names found elsewhere in the British Isles that have been identified as thing sites and which provide additional support for Dingwall’s identification as a thing place-name. These include the comparable Dingieshowe, the possible Dinghill in Leicestershire, Dingbell Hill in Northumberland, and the historic Dingesmere, recently identified with modern Thingwall in the Wirral, England (Baker and Brookes 2015:6; Cavill et al. 2004:29–32; Fellows- Jensen 1996:19, 24–25). At Dingwall the /d/ sound is most likely to have evolved as a result of transmission in a Gaelic linguistic context by which the place-name was later adopted into English, as Fellows- Jensen (1996:24) has argued, though this is unlikely to hold true for all examples given their wide geographical distribution (Cavill et al. 2004:29–32). Dingwall’s association with a Norse assembly site is undisputed based on the place-name evidence. As a Scottish example, it is particularly interesting because the place-name is an outlier on mainland Scotland’s North Sea coast and lies near the presumed southerly extent of Norse influence down the east coast of Scotland (see Crawford and Taylor 2003). Attempts to explain this geographical location have been limited by a paucity of historic information for this part of Northern Scotland during the Viking Age and the limited archaeological information available (see Crawford 1995, Woolf 2010:275–77). This article seeks to address these problems by taking a cross-disciplinary look at where exactly the site of the Dingwall thing may have been located and why. We undertook a detailed documentary study of Dingwall’s local historic landscape aimed at identification of the thing site’s location. The results of this effort have been used to target the first archaeological investigations at the proposed site. We then discuss the archaeological results in combination with a review of other thing sites and the historic and landscape context for Northern Scotland during the Viking Age and Late Norse Period. Historical and Environmental Background to the Thing To date, the archaeology of the Dingwall thing has received limited attention. The precise location of the assembly site was not adequately defined. In this article, the assemblies associated with the Dingwall thing are assumed to have been primarily seasonal meetings of freemen that provided judicial organization for the surrounding region, through the resolution of civil and criminal disputes. In this sense, the thing could have regional or local administrative functions, for instance to resolve property claims for compensation or infringements of taxation. In practice, by the end of Viking Age and into the medieval period, the deliberative functions of the thing could be largely dominated by higher authorities. This model is based on studies of better-documented things in the North Atlantic and Norway. Research in Scandinavia has indicated that such gatherings may have had their origins among Iron Age tribal societies (Brink 2004, Crawford 1987:204–205, Sanmark and Semple 2008, Storli 2010). Since the late 19th century A.D., the main candidate site for the thing at Dingwall was a natural terraced hill called Gallowhill. This feature is located on the western edge of the modern town of Dingwall.1 Mainly based on the apparent judicial association of the name Gallowhill, Bain (1899:45) proposed the location’s connection with the legal functions of a thing, and put forward the unsubstantiated notion that the hill may have been used to seat judges who oversaw Norse courts. Bain’s ideas remained largely unchecked throughout the 20th century. Most recently Crawford (1987:206–208) highlighted the Gallowhill as a candidate site for the thing. The place-name is indeed of interest in so far as it does appear to have a Norse origin. Gallowhill is recorded in a charter of 1603 as Gallibber, which may be derived from Norwegian galgeberg, “the gallow hill”.2 However, this place-name is clearly indicative of a site of capital punishment and so need not correlate with the siting of a judicial court and its associated assembly-field. For instance, a separate gallow-hill and thing site can be seen at Tingwall in Shetland where the court and possible execution site are situated 500 m apart (Coolen 2016 [this volume]).3 This was also the arrangement at estate centers and court-sites in other parts of medieval Scotland and is a pattern identified in Anglo-Saxon England (Innes 1872:97–98, O’Grady 2014:132, Reynolds 2013). We are putting forward an alternative site for the Dingwall thing. The proposed location is a short distance north of Dingwall High Street. This was the site of a mound and level field that were located adjacent to St. Clement’s Church, the medieval parish church. The area has been substantially affected by post-medieval development and land reclamation. A reconstruction of the historic landscape using cartographic and documentary evidence nevertheless shows that the mound was previously located on a small estuarine peninsula. By combining this new landscape analysis with the results from targeted archaeological investigation, we propose that the mound was constructed during the 11th and 13th centuries A.D., in the context of interaction between Gaelic and Late Norse legal traditions. There is a substantial amount of landscape and place-name Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 174 evidence to support the site’s identification, despite extensive post-medieval redevelopment of the immediate environs. Key to this argument will be the notable similarities between Dingwall’s historic landscape characteristics and other better-documented legal assembly sites in Scotland and Norway. Dingwall’s geographical, historical, archaeological, and toponymical context The modern town of Dingwall was developed on the plan of a medieval burgh established in A.D. 1226 and is located in the historic district of Rossshire in northern mainland Scotland (Fig. 1). Nearby to the east is the muddy canalized mouth of the River Peffery, at the point where it meets the head of the tidal Cromarty Firth, which in turn provides access to the North Sea (Fig. 2). The Scottish Gaelic name for Dingwall is Inbhir Pheofharain (Innerpeffery), which translates into English as the “mouth of the River Peffery” (Watson 1904:93). Such a wellappointed location confers on Dingwall a degree of strategic importance. Silting of the river mouth, and to a lesser extent post-glacial rebound, has limited navigable access to the sea since the post-medieval era. The settlement’s strategic importance, at least for the medieval era, is illustrated by Dingwall Castle’s representation as one of the main fortifications in northern Scotland on Matthew Paris’s ca. A.D. 1250 map of Britain.4 Partial remains of the castle are still visible in the grounds of a 19th-century A.D. mansion near the mouth of the River Pef fery.5 The settlement is located at the eastern end of the agriculturally fertile Strathpeffer valley and at a short distance north from the mouth of the Canon River, the largest river in the district (Fig. 2). Overland routes to the east provide access to the Black Figure 2. Map of Dingwall showing places mentioned in the text. (Background map reproduced from Ordnance Survey map data by permission of Ordnance Survey. © Crown Copyright 2016.) Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 175 Isle peninsula, which divides the Cromarty Firth from the larger Beauly Firth and the historic settlement of Inverness, where the district of Moray lies beyond to the southeast. Landward routes are also available via Strathpeffer to the west coast of Scotland by way of Strath Bran and Glen Carron. This geographical location with its protected sea loch (lake), good access to the North Sea, and accessibility to natural resources would have made this an ideal location for Norse expansionist interests in the area from the Viking Age onwards. Evidence for Viking activity. Norse interest in the area is supported by the occurrence of Norse place-names there, including Stava, Stavek (1266 A.D.), and Stavaig. These are obsolete historic names for the lower tidal reaches of the Canon River and were derived from ON stafr-a “stave river” and ON stafr-vik “stave bay” (Crawford and Taylor 2003:8; Macrae 1923:152; Watson 1904:xviii–xix, 152). Partially based on this toponymic evidence, Crawford (1986, 1995) has proposed that the area was targeted for Norse settlement from the Northern Islands of Scotland during the 11th century A.D., with the specific purpose of exploiting local timber resources for trade and ship-building. In support of the feasibility of this idea, there are notices of timber shipments from vessels berthed near the shoreline at Dingwall recorded as late as 1813. 6 Strathpeffer was also clearly an important center of power in the Pictish era prior to the Viking Age. A class I Pictish symbol stone is located at the Parish Church of Dingwall. It was discovered reused as masonry in 1875 when the church was rebuilt. The church site dates back to at least the medieval period, although the Pictish stone raises the possibility for an early medieval foundation. Perhaps more significantly for the present discussion, the church’s dedication was to St. Clement’s, which may imply a Norse or Danish connection with the 11th-century A.D. seafarer’s cult (Crawford 2008:199–200). Further to the west and overlooking Dingwall are the remains of a hillfort at Knockfarrel. This fort was most probably the main power center in the area until the Viking Age, and standing nearby in the valley below is another class I Pictish symbol stone.7 Following Woolf’s (2006) reorientation in our understanding of the political geography of Pictland, it has now become apparent that Dingwall and Ross are likely to have been located within the important Pictish kingdom of Fortriu until the late 9 th century A.D. This association has relevance for the current discussion because among the earliest mentions of Viking activity that were likely to have affected the area around Dingwall is an entry in the Annals of Ulster for the year A.D. 839 that recorded a decisive military defeat of the men of Fortriu and their allies by “the heathens”.8 The apparent scale of this event may indicate the action of a large Viking military campaign rather than a seasonal expeditionary force (Woolf 2010:66). Although caused by unknown antagonists, this battle appears to have been a catastrophic event for the political map of northern Britain, as Fortriu recedes in our sources from this date. This event may also provide historical context for the establishment of a head thing in the region by Viking settlers, marking the reorganization of legal traditions and subjugation of the local population. There are ample analogies to draw on for similar large-scale military campaigns by Danish forces in Anglo-Saxon England during the 9th century A.D., which in time affected the organization of legal assemblies. 9 In contrast, a smaller-scale event is recorded in the north for A.D. 866. The Irish annals record a Hiberno-Norse invasion of Fortriu and Pictland under the leadership of Amlaíb, son of the king of Laithlind, and his brother Auisle, which only appears to have lasted three winters (Anderson 1973:250).10 In the following years, possibly the same Amlaíb was active in several other locations in the British Isles and had perhaps been killed by the early 870s (Woolf 2010:107–109). The events of A.D. 866 seem unlikely to have involved the managed creation of a stable legal assembly site at Dingwall, though of course the thing may have been convened before A.D. 866. By the early 10th century A.D., the political center of Pictland had shifted southward, away from the areas later known as Moray and Ross, to the vicinity of Strathtay and modern Perthshire. A general consensus has emerged among scholars that this change was in part due to damage inflicted upon Fortriu by Viking incursions during the 9th century A.D. (see Woolf 2010:87–121). As with the western Atlantic seaboard, the extent of Viking settlement and political dominance established north of the Moray Firth is difficult to prove, though it has obvious implications for our understanding of the extent to which new Norse legal practices and assembly sites were also introduced. There is only limited Viking archaeology in the form of material culture and burials from Ross and neighboring regions. No in situ Viking burials are known from the area. Nonetheless a small, but significant group of artifacts have been discovered in the north of the region near the Dornoch Firth, 70 km northeast of Dingwall. This collection includes an oval brooch found with a fragment of steatite bowl at Ospisdale (Curle et al. 1954:238–239, PSAS 1933:20); a fragment of hack-silver from the terminal of a ring-money bracelet found at Pitgrudy (Hunter 2011:104); and a late-10th-century A.D. silver hoard from Portmahomack (Miller and MacLeod Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 176 1889:314–317). Recent excavations at the associated early medieval monastery at Portmahomack have also revealed a Viking Age destruction horizon dated to the late 8th century or early 9th century A.D. (Carver 2008:136–139). Further north at Dunrobin in Sutherland, discovery of an iron axe head, a ring, and two oval brooch bases may indicate the site of another disturbed grave or graves. Further south, a 9th-century A.D. copper-alloy Hiberno-Norse ring pin was recently discovered by metal detecting near Castle Stuart on the Moray Firth, and a possible Viking horse burial is known from the Moray–Banff border (Graham-Campbell 2004:216, 227).11 As a group, these finds amount to a tantalizing, though fairly limited, snapshot of Viking activity across this extensive coastal region at the height of the Viking Age; however, it is noteworthy that no finds are known near Dingwall. Despite this paucity of material information, other types of evidence do suggest that Norse settlement had extended further south to the Cromarty Firth. Place-names. More generalized information about Norse settlement in the vicinity of Dingwall can be established from ON place-names (Crawford and Taylor 2003, Fraser 1986, Watson 1904). Near Dingwall, westward up Strathpeffer, is Ulladale, which is from Ulli-dalr12 with the meaning “Ulli’s valley”—presumably referring to a local Norse freeman or tenant (Crawford and Taylor 2003:8, Watson 1904:100). And elsewhere, Scatwell13, in Contin Parish ~14 km west of Dingwall, was ON skat-völlr, “tribute or tax field” (Crawford 1986:34; Crawford and Taylor 2003:8–9, 29; Watson 1904:149). According to Crawford (1986:43), “scat” was a tax that in Scotland “was exclusively associated with the Dominion of the Earldom of Orkney”. Swordale14 to the north of Dingwall in Kiltearn parish is ON svörð-dalr meaning “grassy sward valley”, and nearby Katewell15 was ON kví-dalr, “sheep-pen valley” (Crawford and Taylor 2003:8, Watson 1904:87). Notably, the Cromarty Firth region and neighboring Black Isle demarcate the most southerly extent of ON dalr and bólstaðr (“settlement, farm”) with a single outlier in Morayshire to the southeast. These place-names are compelling evidence for local Norse nomenclature and indicate a social context of dispersed settlement that the Dingwall thing would have served. Nevertheless, Crawford (1986:43) and Grant (2000:95–96), among others, have highlighted the apparent lack of place-names for ON ouncelands or pennylands. These terms were land-valuation units on which payment of tax or tribute was calculated and levied. They are found in most other areas of Scandinavian settlement in northern and western Scotland, which Williams has argued resulted from an expansion of the authority of earls from Orkney out from the Northern Isles, under the political influence of earls Sigurd Hlodversson and his son Thorfinnr the Mighty (Williams and Bibire 2004:203–204). The absence of ouncelands and pennylands from Ross may indicate that earlier Norse settlement had been integrated into the indigenous system of davoch land divisions. Davochs are prevalent throughout Ross and neighboring Moray and are closely associated with expansion of the Kingdom of Alba and neighboring Morayshire (Ross 2011:14–33). Crawford and Taylor (2003:10) have alternatively reasoned that a semi-autonomous colony is suggested by the pattern of Norse settlement names in Ross, perhaps established under the auspices of the Norwegian exile Kalf Arneson who fled to Earl Thorfinn in A.D. 1035. The lack of Late Norse land-division systems may also reflect the less successful expansion of Norse authority into Ross compared with Caithness, where the dominance of the earls and later the kings of Norway was established on a more stable and long-term basis. In this way, Dingwall should perhaps be seen more as a changeable border zone of Norse influence than part of a fully settled area of control. The thing place-name does, however, suggest the presence of a gathering place used to resolve disputes among these communities, however autonomous or otherwise they may have been from comital or royal authority. Other place-names near Dingwall provide further topographic information about the assembly site’s immediate environs. An example is the place-name Broadpool, which is recorded as the northern boundary of Walkman, a large open field, in a Burgh sasine register for Dingwall dated to the late 17th century A.D.16 This name contains the element pool with the meaning “harbor” or “shore”, which is explicable as a derivation of ON pollr, “a pool”, specifically one which at ebb of tide retained water sufficient to float a ship. The generic component of Broad or Brod, may derive from ON brott, braut, or more particularly Shetland Norn brød or bröd, each bearing that same meaning of a made way or track. This posterior positive word-order displayed in the place-name Broadpool is common in Gaelic, but it was also common in medieval Norwegian into the 14th century A.D. Alongside the Broadpool boundary ran a ditch parallel to it on its northern flank known as the stripe of Broadpool. Taken together this implies Broadpool may have meant “the way or road leading to the shore”, and potentially was a track engineered by Norse settlers through wet carseland to reach a berthing location on the Canon estuary previously known as the Stavek (see above). 17 The place-name Gallowhill at Dingwall has already been mentioned. It is recorded in a charter of 1603 as Gallibber and in burgh sasine register Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 177 and Moray and more than half of Scotland”. This Sigurðr has been identified with one “Sichfrith the Earl” mentioned in an entry for A.D. 893 in the Annals of Ulster and was apparently active in Dublin (Benediktsson 1986, cf. Woolf 2010:284).20 The accuracy of the saga’s account of the annexing of Northern Scotland has been called into question due to problems in the lineage of individuals mentioned in the narrative and other cross-referencing errors within the texts (ibid). A reference in the Chronicle of Kings of Alba to the “wasting of Pictland” may, however, provide some historical context. This event occurred during the reign of Domnall mac Constantín (A.D. 889 x A.D. 900) who was killed by “heathens” at Dunnottar in Aberdeenshire, which might coincide with the supposed invasion by Sigurðr and Thorsteinn (Hudson 1998, cf. translation in Woolf 2010:122). Sigurðr is said to have died near Oykel in A.D. ca. 892 on the Dornoch Firth. A mound at Cyderhall (ON Sigurðar-haugr) has been proposed as his burial site (Crawford 1986:38–40). It may be significant that this is near to the placename Cuthill Links, which is from Early Gaelic comhadhail, meaning “place of assembly” (Barrow 1992:231, no. 1.1; O’Grady 2014:115). In the Orkneyinga Saga, Thorfinnr Earl of Orkney and son of Sigurðr Hlodvisson is also said to have been active in northern Scotland. His exploits apparently included holding the lordship of Caithness and successfully invading the north of Scotland, culminating in the unidentified battle of Torfness against Karli Hundason. Karli has been identified as King MacBethad mac Findláich of Alba—the historical Macbeth (Cowan 1993:125– 126, Crawford 1987:72). However, many of Thorfinn’s supposed exploits as described in the sagas are not corroborated by contemporary source material, and so the relevance of the saga’s narrative to the history of northern Scotland during the Late Norse period is contested (Woolf 2010:243–244, 309–310; cf. Crawford 1995, Ross 2011:124). Woolf has argued that Thorfinn’s legacy had been conflated with a later earl’s career to enhance his credentials as an eponymous ancestor for the Orcadian rulers of the 12th and 13th centuries A.D., particularly Haraldr Maddaðarson who occupied much of northern Scotland in the A.D. 1190s (McDonald 2003:39–40; Woolf 2010:243–244, 283). With similar pessimistic caveats, Njál’s Saga written in ca. A.D. 1280 and Orkneyinga Saga describe an earlier battle in northern Scotland between Earl Sigurðr digri (Thorfinn’s father) and two Scottish earls connected with Moray, although the accounts do not agree on specific details, including the location (see Ross 2011:125–126). However we choose to treat the veracity of conflicts mentioned in the sagas, it is entries it is variously named Gallowber of Gallowhill (1684), Gallober (1699), Gallabir (1729), Galloper (1742), and Galliper (1805). As discussed, this place-name in its various recorded forms may be derived from medieval Norwegian galgeberg, “gallows’ hill”, indicating the pairing of an execution site with the nearby court meeting place. Gallowhill is ~900 m west of both the medieval burgh center and the location that we are proposing was a thing site. A further alleged judicial execution site that may have early medieval pedigree can be identified 34 km south of Dingwall at Tom Na Croiseige by Kiltarlity. Here a stepped-mound 23 m in diameter and 3 m high is traditionally said to have been a “seat of judgment” where a “hanging tree” grew.18 The functionality of stepped-mounds as assembly sites has long been suspected based on the well-documented example at Tynwald on the Isle of Man, with other examples at Thingmote in Dublin, Law Ting in Cumbria (England), and—less certain—Tinwald in Dumfriesshire and Govan, both in Scotland (Cowper 1891; Darvill 2004; FitzPatrick 2004:28, 45–47; O’Grady 2008a:211–217; Owen and Driscoll 2011:341). The execution association, however, may not necessarily indicate a court site, and the age of the Kiltarlity mound’s steps is not known. Still, it may be significant that another stepped-mound is located a short distance west of Dingwall at Fodderty, the site of the medieval parish church of Fodderty (NGR NH 5130 5936). Either one of these stepped-mounds could potentially have been a thing mound in the Hiberno– Norse Irish Sea tradition, but in the absence of concerted archaeological investigations at these types of site, beyond Tynwald on Man, it is at present difficult to advance such a case any further.19 The sagas. Exactly when Norse control may have extended to the area later known as Ross has been a point of considerable contention, not least because of our reliance on problematic sagas to give anything like a narrative history for the area during the Viking Age and Late Norse Period. There is notoriously little agreement about the historical value of saga accounts relating to northern Scotland, such as in the Landnámabók, Orkneyinga Saga, and Njál’s Saga, texts that were mainly composed by Icelandic authors between the late 12th and 14th centuries A.D. (Woolf 2010:277). The main protagonists whose careers are linked to northern mainland Scotland were Sigurðr the Mighty, son of Eysteinn Glumra, and Thorsteinn the Red in the late 9th century A.D., and Thorfinnr the Mighty, son of Sigurðr Hlodvisson, who were both Earls of Orkney during the first half of the 11th century A.D. Sigurðr son of Eysteinn Glumra and Thorsteinn were said in Landnámabók to have conquered “Caithness and Sutherland, Ross Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 178 mormaer of Moray, as King of Alba who died in A.D. 1057, and possibly the career of Mael Snechta mac Lulaich who was referred to as rí Mureb (“king of Moray”) in his A.D. 1085 death notice.21 In this section we have aimed to provide an outline of the geographical, historical, archaeological, and toponymical context for Dingwall during the Viking Age and Late Norse Period. Two key historical periods have emerged as potentially important for interpreting the development of the Dingwall thing: the 9th century A.D. and the early 11th century A.D. The information considered above will form the basis from which to better understand the historic significance of the thing site’s location and physical attributes when we turn to the archaeological findings. At this point, we will explore the detailed local historic landscape evidence that has aided identification of the thing site. Identification of the Thing Site In A.D. 1226, Alexander II chartered Dingwall “in Ros” as a royal burgh; the medieval place-name, as we have mentioned, indicated the prior meeting place of a thing. Given the ON element –völlr (“a level field”), the location of the common gathering of the thing is likely to have been on the level valley floor visible at the lower floodplain of the River Peffery. The “field” element is commonly associated with thing sites in the British Isles and North Atlantic (Coolen and Mehler 2014, Crawford 1987:204– 210, Darvill 2004, Fellows-Jensen 1993:55–56). A usual feature is for a core meeting-place, where court proceedings were convened, to be located at or beside the assembly field.22 In this section of the article, we set out a case for the specific location of the Dingwall thing site and advance the theory that the site correlates with a later medieval assembly mound, which we argue had reused the venue for the earlier thing. The Mute hill of Dingwall was a large earth mound that appears to have been a medieval judicial meeting place, and was previously located on the south bank of the River Peffery, a short distance from Dingwall high street. The site’s topography and landscape can be reconstructed using localized historical and cartographic evidence. This reconstruction will show that the site had striking similarities with other better-known thing sites. Identification of this site with the thing is not a new notion, as Bain (1899:44, 103) and Watson (1904:93), and more recently Fellows-Jensen (1996:24), suggested that the Mute hill of Dingwall could have been significant to the thing. However, the precise location of Mute hill had not been proven by previous commentators, and no thorough study of this historic feature had been attempted (cf. Clark 1993). apparent from reconstructed genealogies that substantial integration had occurred between the ruling families of the Northern Islands and mainland Scotland. Such political integration may have extended to other forms of integration in legal culture and mechanisms of lordship, a subject that we will return to below. Based on the accounts and historic references outlined above, it is probable that the north of Scotland had come under some form of Norse rule as far south as Ross at some point between the late 9th and early 11th century A.D., even if it is not possible to reconstruct the precise longevity and details. Late Norse period and medieval historical records. Although fragmented, medieval records from the Scottish perspective can also relate useful information about the political status of Ross and neighboring Moray during the 11th and early 12th century. These accounts have implications for understanding Dingwall’s importance in the Scottish Late Norse Period. Ross was clearly a province in its own right at least as early as ca. A.D. 1115 when an earl Aedh appears as a witness to Alexander I’s founding charter of the priory of Scone. The same earl Aedh was witness in the A.D. 1120s to two early charters of David I (Lawrie 1905:nos. 36, 74, 94). In 1157, Malcolm MacHeth was recognized as earl of Ross by Malcolm IV, after a period of royal confiscation that followed a rebellion led by the mormaer (equivalent to an earl) of Moray in 1130 (Innes 1840:151). Records of a subsequent rebellion in Ross by the MacWilliams during the 13th century A.D. include mention of the “thanes of Ross” in 1211 (Skene 1871:278, ii, 274). Grant (2000) has made a case for the existence of a thanage at Dingwall from the early 11th century A.D. If true, this thanage may have been formed to subdue the region after annexation and to create a bulwark of royal authority against future incursions from the north. Much discussion has also considered the question of the political status of neighboring Moray during the 10th and 11th centuries A.D., with debate centering on whether Moray can be defined as an independent kingdom at some point during the 11th century A.D. This is not the place to debate further whether it was mormaers or independent kings based in Moray (Moréb) that held sway over large swathes of northern mainland Scotland between Caithness and the Cairngorms mountain range (Dumville 1997:36; Woolf 2000, 2010:240–241; cf. Ross 2011:82–100). Nonetheless, what is relevant for the present discussion is that lords in Moray would certainly have had a vested interest in maintaining control of, or nullifying, a regional judicial center in the neighboring province to the north; either by proxy or directly. Key episodes that may have affected Dingwall include the rise of MacBethad mac Findláich the Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 179 Relevant here is Fellows-Jensen’s (1993:64–65) conclusion that the majority of thing place-names in Britain indicated sites where “legal assemblies continued to be held for centuries, even after the Scandinavian language dropped out of use”. Conversely, recent study of the locations for local assemblies in Viking Age and medieval Norway have highlighted the potential for change and complexity in the siting of local and regional courts (Ødegaard 2013). On this basis, it should also be borne in mind that the siting of the thing at Dingwall may have changed over time within the general environs. Beyond the firm place-name evidence for the thing reviewed above, other evidence for a historic judicial meeting place at Dingwall comes from later medieval historical documents. When in 1503 James Duke of Ross resigned his lands of Ross to his brother King James IV, it was recorded that in order to keep his title the Duke retained the montem of Dingwall, from the Latin meaning “mound, a small hill”, that was described as juxta (“beside, near”) the town (Dalrymple 1770:58).23 The mound referred to in the 1503 record was not the castle mound of Dingwall, which was located further to the east and was a large stone-built fortification occupied up until A.D. 1625 (NMRS site no NH55NE 4, RCAHMS 1979:30, no. 255). The only recognizable mound that can be identified as having stood adjacent to the medieval burgh was a low mound that previously stood within a walled-enclosure a short distance north of the high street. This mound was used for the burial site of George Mackenzie, first Earl of Cromartie, who died in A.D. 1714. Since A.D. 1947, the enclosed area and site of the by then levelled mound have served as a town center car park. The title deed gained by the first Earl of Cromartie to what would be his place of burial was dated February 13th, A.D. 1672 and entitled “… disposition of ye mute hill of Dingll to Tarbatt”; Tarbatt in this case referred to Sir George Mackenzie who was also Viscount Tarbat.24 Mute hill is cognate with “Moot-hill” and is Scots for “hill of assembly”; derived from Old English mōt or gemōt with the meaning “assembly”, “meeting”, or “encounter”. Place-names in the form mute-hill have been revealed as a significant indicator for medieval judicial assembly sites in Scotland, some of which may have origins in the late first millennium A.D. (O’Grady 2014:129–134). The earl’s A.D. 1672 property disposition refers to the area within which the mound was located as the Hillyard, which indicated an enclosure surrounding the Mute hill. The precise location of this enclosure is described in the document as delineated by the churchyard of Dingwall parish church on the north and at the west by the “the co’on Calsay that leades to the church yaird of dingll”.25 Calsay is Scots for a stretch of paving or a paved part of a street.26 On the east, the Hillyard was bounded by slaik, common burgh land. Slaik is Scots for “to lick”, which in this context refers to the licking or lapping movement of the tide and describes land which the sea covers at high water, in this case mudflat.27 John Bayne, minister of Dingwall, in A.D. 1718 referred to a “ditch or sink” which “wants a bridge … betwixt the church and the glebe”.28 Richard Pococke, a Church of Ireland Bishop for Ossory, noted from his visit to Dingwall in A.D. 1760 that “to the South of the Church is a stone enclosure in ruins but fenced with a Ditch which is the burial place of the family of Cromartie” (Pococke 1887). Land records of the 17th and 18th centuries A.D. indicate that the northern boundaries of the burgh tenements of Dingwall appear as “the slaik” or alternatively “the sea, the floodmark, or the fludder” (cf. fløda, Shetland Norn, “reaching high water”), except those which backed on to the Hillyard assembly site.29 In 1684, “slaick” formed the northern boundary of the Trinity Croft. This feature is a large open field which is mentioned with the Hilyaird in a royal charter of 1591.30 It was one item amongst chaplainry lands of the Holy Trinity and St. Michael in the burgh of Dingwall, the superiority of which the crown granted in 1591 to one Ronald Bayne, the burgh’s commissioner to parliament. In the charter of demission, the “Hilyaird” is recorded as “horto”, a yard or enclosure, pertaining to the “Trinitie-croft”.31 The Trinity Croft is indicated on the Ordnance Survey map of A.D. 1881 as “Trinity”, a field to the immediate west of both Church Street and the parish churchyard. The name may come from the adjacent parish church, one of the dedications of which is to the Holy Trinity. This legal attachment of the Trinity Croft to the Hillyard suggests that historically the two had been components of a single entity. Although the evidence is admittedly late in date, the Trinity Croft, with its level topography and legally defined boundaries, is an attractive candidate for an assembly field associated with the court mound. Also of relevance is a burgh sesine record of A.D. 1682, which records that at least one of the chaplainry properties associated with Dingwall, Aikermichael, had been held by the abbey of Dunfermline.32 Dingwall Parish Church, located beside the Trinity Croft, had also remained a property of the Priory of Urquhart in Moray until shortly before 1500. Urquhart was a Benedictine sister house of Dunfermline. The chaplainry properties and the church may have been part of extensive lands bestowed on these monasteries by David I in A.D. 1135, as part of a reassertion of royal power in the district of Ross, which followed the defeat of Oengus mormaer of Moray at Stracathro in A.D. 1130 (see Ross Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 180 forming a variety of small islands and peninsulas beside the town (Rose 1791:4). A representation of this environment can be seen on two maps created by George Brown in the A.D. 1790s (Fig. 3).33 The Hillyard is visible as an enclosed area south of the parish church. On each of Brown’s plans of Dingwall is shown an island in the tidal water east of both the church and Hillyard. In a charter of 1547, that island is recorded as “Cruke, lying east of the church”, and on David Aitken’s plan of “The Burgh Lands of the Estate of Tulloch”, dated A.D. 1789, it is named as Easter Cruik.34 In burgh records, the island is variously recorded as Cruik, Little Cruik, or Easter Cruik, and said to be bounded on all sides by “the sea”. Cruik in Scots means a bend (cf. ON krokr, “bend”). The name no doubt refers to the pronounced southern bend in the river that enclosed the island and which formed a tidal inlet beside the 2011:78). Even by 1157 when Malcolm IV reinstated Malcolm MacHeth as earl of Ross, the earl was ordered to “protect the monks of Dunfermline and all that was theirs in the time of King David” (Barrow 1960:222, no. 179). On this basis, royal confiscation of the assembly site at Dingwall may have been a conscious part in this political strategy to effectively nullify the judicial center that would have been an important apparatus of lordship in the region. This could also mean that the site had a ceremonial importance for the mormaers of Ross in the early 12th century A.D., perhaps as an inauguration site? The historic landscape of the site can be further reconstructed by a reference in A.D. 1787 to their having also been “Slake … at the east and north of the Church” (Minutes of Dingwall Town Council, 2 October, A.D. 1787). In addition, in 1791 the lower tidal waters of the river were described as Figure 3. Extract from a plan of intended road from Lochcarron to Achnasheen and thence to Dingwall surveyed in 1793 by George Brown. © Crown Copyright: National Records of Scotland (Ref. No. RHP 11597 and 11591). Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 181 mound site. Based on this evidence, the Mute hill mound should be understood to have occupied a small peninsula, which extended northwards into the lower tidal waters of the River Peffery. From the early 19th century A.D. onwards, this historic estuarine landscape was progressively obscured. This process began with the canalization of the river under the direction of Thomas Telford during A.D. 1815–1817. The resulting changes and reclamation of land can be seen on A.D. 1821 maps by Telford and John Wood (Figs. 4, 5). On Wood’s map, the mound is visible within the Hillyard enclosure south of the parish churchyard, a short distance north of the town. The Trinity Croft field is a blank area immediately west of the mound across Church Street. An early photograph from the A.D. 1880s shows the Trinity Croft still in Figure 5. Plan of the Town of Dingwall, surveyed by John Wood in 1821. The Mute hill mound is visible labelled as the “Earl of Cromarty’s Memorial”. © National Library of Scotland. Figure 4. Extract from Telford’s plan of Dingwall Canal. Commission for Highland Roads and Bridges Eighth Report 1821. Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 182 use for farming beside the church of St. Clement’s. The mound is out of view behind buildings to the right (Fig. 6). This picture again illustrates the close proximity of the historic mound, church, and field. This relationship is illustrated clearly on another late 19th-century A.D. map of the area (Fig. 7). The mound was the subject of an antiquarian excavation in A.D. 1875, which sought to confirm Figure 6. The Trinity Croft in the A.D. 1880s. Courtesy Dingwall Museum Trust. Figure 7. Extract from Ordnance Survey 1st edition 6-inch map of Ross-shire and Cromartyshire, Sheet LXXXVIII, surveyed A.D. 1873 and published A.D. 1881. © National Library of Scotland. Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 183 the exact location of the Earl of Cromartie’s burial place, a basic account of which was recorded by Fraser (1876:clxxv). The excavation does not appear to have been undertaken with awareness that the site had once been an assembly mound. Based on Fraser’s description of the excavation, this was an unscientific and exploratory event. Nonetheless, the drawings that Fraser made of the Earl’s 18th-century tomb are valuable as the only archaeological record of the mound prior to its levelling, though these reveal little about the mound’s composition or appearance (Fig. 8). Two photographs from the early 20th century show the mound prior to its levelling in A.D. 1947 (Fig. 9). These show a low turf-covered mound with gradual slopes spread within the Hillyard’s enclosure wall. After World War II, the mound site was used as a town center car park and renamed the Cromartie Memorial Car Park. This involved levelling of the mound in A.D. 1947 using a bulldozer, and resurfacing of the site with tarmac (Figs. 10, 11).35 The neighboring Trinity Croft field was used as the site of an agricultural auction market for much of the 20th century until 2003 when redevelopment of the site for a supermarket and petrol station took place. Prior to this, archaeological works for development management purposes took place on Church Street and at the site of the former auction market. Twelve trenches 1250 m2 in area were excavated across the site of Trinity Croft up to the western border of Church Street, but no archaeological deposits were encountered; the natural substrate was confirmed as a sticky grey clay (Cook 2003). If the Trinity field was indeed the site of the popular thing assembly, unfortunately it appears unlikely that significant archaeological remains have survived in this area. Despite this result, there was still potential for the mound site to contain significant archaeological remains. New archaeological investigations directed by Oliver O’Grady were therefore targeted at the Cromartie Memorial Car Park. As this section has detailed, this location can now be appreciated as the site of the historic Mute hill of Dingwall, and its historic landscape setting on an estuarine peninsula, with the adjacent field and church, is a strong candidate for the central meeting place of the thing site. Details from the results of the new archaeological investigations will now be described. Archaeological Investigation The investigations were undertaken during 2011–2012. This project began with a ground penetrating radar (GPR) survey that was aimed at assessing the extent to which archaeological remains of the mound and adjacent estuarine shore had survived beneath the car park. The survey helped to confirm the presence of extensive and complex archaeological remains, including heavily truncated basal deposits of the mound; as well as Figure 8. Illustrations from an excavation at the Earl of Cromartie’s burial place on the Mute hill in 1875 (Fraser 1876:clxxiv–clxxv). Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 184 post-medieval building remains, in particular on the southern side of the site. The location of the historic shoreline of the peninsula on which the mound was positioned was also revealed. Based on these results, a single trench was excavated across the north side of the mound. This was intended as a means of verifying the geophysical results, but also sought to recover material culture and dating evidence that Figure 9. Photograph (top) taken from the north in the early 20th century A.D. that shows the Mute hill mound in the Hillyard with the town behind. Photograph (bottom) taken from the southeast during demolition of the mound in 1947. Courtesy Dingwall Museum Trust. Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 185 Figure 10. The Cromartie Memorial Car Park at Dingwall viewed from the south. Note the low rise in the car park’s topography that indicates the mound site. © Oliver J.T. O’Grady. Figure 11. Topographic contour plan of the mound site (10-cm–contour separa tion). Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 186 Figure 12. Interpretive plan of radar timeslice anomalies. could help understand the function and chronological development of the mound. Although no Viking Age material culture was recovered, an important sequence of radiocarbon dates support that the mound was a man-made monument and created around the mid-11th century A.D. during northern Scotland’s Late Norse Period. The excavation also revealed that the mound was constructed from redeposited clays during at least two, potentially three, separate phases of construction, the first around the mid-11th century and latter possibly during the 12th–13th century. We first describe the results from the geophysical survey and then provide a summary of the excavation findings. Geophysical survey The GPR survey was undertaken across the area of the car park and over adjacent road surfaces at Church Street and Tulloch Street. The equipment used was an Utsi GV3 single channel 400 MHz central frequency antenna. A timeslice survey was recorded across the entire interior of the car park using a line separation of 0.5 m, sample resolution of 14.75 mm, and a 60 ns time sweep. Single radar profiles were recorded to the north and west of the car park (O’Grady 2011a:5–6).36 The timeslice survey revealed the historic extent of the mound and the reclaimed estuarine shoreline as a strip of homogeneous disturbance around the north and east side of the car park (Figs. 12, 13). The base of the mound was registered as a coherent oval-shaped zone of interfaces approximately 45 m north to south and 39 m west to east (Figs. 12, 13). Anomalies 10 m–15 m wide across the north and south of the car park were interpreted as areas of reclaimed tidal inlets. This finding suggestJournal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 187 Figure 13. GPR timeslice plots from the mound site. ed that a channel up to 3 m deep bordered the site to the north. The perimeter of the peninsula shoreline appeared uneven. This may reflect the historic layout, but areas of possible truncation and ground disturbance particularly at the north and northeast may also account for the irregular plan (Figs. 12, 13). Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 188 More substantial disturbance was evident to the south. This area correlated with the location of buildings that stood here during the 19th century A.D. (Figs. 12, 13). Modern service trenches were also evident as were areas of disturbance probably related to reduction of the mound in 1947 (Figs. 12, 13). Part of a possible causeway or neck of land that connected the peninsula to the area of Trinity field was evident at depth as a distinct zone of interfaces on the southwest side of the site (Figs. 12, 13). Several anomalies across the body of the mound indicated structures and areas of disturbance. For instance, a large rectangular anomaly indicated the site of a stone wall that previously enclosed the center of the mound (Figs. 12, 13). This anomaly, which was later confirmed by excavation, relates to an 18th-century A.D. burial enclosure. Further post-medieval burial activity and Victorian excavations were indicated by irregular areas of disturbance around a modern memorial garden that now occupies the center of the mound. The garden was not surveyed due to the likelihood of extensive modern disturbance and the limited available space in this area (Figs. 12, 13; cf. Fraser 1876:clxxv). A single radar profile recorded across the west side of the site provided a complete section through the in-filled tidal inlet that previously surrounded the side of the mound (Fig. 14). The radar results informed the positioning of an excavation trench across the northwest side of the site, where modern disturbance appeared least, and the results of which will now be described. Excavated evidence A trench 2 m wide and 20 m long was opened across the northwest side of the car park (Fig. 15). This excavation aimed to characterize the mound’s composition, the stratigraphic relationship with the old shoreline and peninsula (Fig. 16), and to investigate a series of smaller anomalies identified by the radar. A summary of the excavated sequence will now be given (cf. O’Grady 2012). 37 Modern layers of made-up ground were revealed beneath the tarmac and overlay the remnant slope of the mound (Fig. 17). Other post-medieval features included remains of the Hillyard’s enclosure wall and the foundations of the modern memorial garden wall at the southeast end of the trench. This wall was cut into the clay of the mound. A rectangular enclosure identified in the radar survey was revealed as a mortared section of wall-footings across the middle of the trench (Fig. 18). This feature appears to have been an enclosure wall and surrounded the Earl of Cromartie’s burial memorial that was built into the mound around A.D. 1714 (O’Grady 2012:5–7). At the northwest side of the trench, estuarine muds were revealed as a purplish dark brown clay containing sand, mollusc shells, charcoal flecks, post-medieval pottery, and animal bone. The mud was explored to a depth of 1.5 m, though the base was not located. To the southeast, near the mound, this deposit overlay layers of sandy clays and was interpreted as the last estuarine muds to have been laid down by the action of tidal water prior to land reclamation during the 19th century A.D. (010; Figs. 19, 20). A slot was excavated through the interface between the estuarine muds and the edge of the mound (Fig. 21). The outer slope of the mound overlays a further estuarine mud that in turn itself overlays an earlier berm of the mound. Northwest of this slope, the estuarine mud went down onto a compact peat (O’Grady 2012:7–8). This deposit of peat with clay lenses sloped down to the northwest and was shown Figure 14. GPR profile T6. Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 189 to overlay the lower edge of a light-brownish grey sandy clay, which appeared to form part of the lower deposits of the mound (Fig. 21). Radiocarbon dates from the upper and lower mound layers indicated that the intervening estuarine mud was deposited between the 11th and 13th century A.D. (Fig. 22). This finding was supported by two redware medieval pottery sherds from the intervening mud that date to the late 12th or 13th century A.D. (O’Grady 2012:49; G. Haggarty, National Museum of Scotland, Edinburgh, UK, pers. comm.). This is crucial evidence that the last phases of the Mute hill’s construction occurred during the medieval period and may relate to a later remodelling of the assembly mound. The 18th-century A.D. enclosure wall mentioned above truncated the upper strata of the mound and in doing so had created an artificial division between the stratigraphy to the southeast and northwest (Fig. 23). To the north side of the wall, the truncated remains of the mound’s outer edge were visible as a low slope. This deposit was a compacted band of dark-brown silty clay (012; Fig. 23). Medieval pottery was recovered from the surface of the deposit and a fragment from the rim of an iron vessel. This layer was interpreted as the man-made remains of the outer surface of the mound. This feature may have formed a berm around the edge of the mound, but more likely has simply been truncated by modern demolition and originally covered the entire surface. Medieval pottery from the underlying deposits and radiocarbon dates (SUERC-45299 Corylus, SUERC-45300 Figure 16. Stratigraphic matrix. Figure 15. Plan of trench location. Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 190 Corylus) indicate that this outer level was created no earlier than the late 12th century A.D. and no later than the 17th century A.D. The layer of estuarine mud beneath the outer mound deposit extended a short level distance beyond it to the northwest giving the impression of a small step (Fig. 23). This feature probably marked the limits of the tidal shoreline around the mound (O’Grady 2012:8). An earlier band of the mound’s construction material was revealed as a bluish-grey clay. This deposit was truncated by the 18th-century A.D. enclosure wall and formed a compacted layer containing flecks of charcoal, exposed to a depth of 0.49 m. Beneath this, a firm light-brownish-grey sandy clay formed a thick band containing patches of ironoxide and charcoal flecks. Radiocarbon dates from Figure 17. Southwest-facing section showing demolition layer (0 02) above estuary mud (010). Figure 18. Pre-excavation photograph of wall [004] looking nort heast. Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 191 the interface of the outer mound deposits and bands of layers in the mound beneath the cut of the postmedieval enclosure wall indicate that this part of the mound was created at some point before 1029–1220 cal AD (95.4%, SUERC-45300, Corylus) and after 980–1164 cal AD, though this date was derived from a long-lived tree species (95.4%, SUERC-45298, Pinus; Fig. 24; O’Grady 2012:8–9). Two further bands within the mound, in the form of dark grey clay, were revealed beneath the post-medieval enclosure wall cut. These deposits sloped down to the northwest and appeared to curve round from the north to the southwest (021 and 022; Figs. 23, 25). Charcoal retrieved from these bands returned calibrated radiocarbon dates of 1046–1275 cal AD (95.4%, SUERC-45297, Corylus) and 980– 1164 cal AD (95.4%, SUERC-45298, Pinus; Figs. 22, 24). The comparatively thin width of these bands may suggest that they formed historic surface material for a smaller mound, earlier in date, though no turf-lines were evident. On this basis, the overlaying thicker mound deposits could potentially indicate Figure 19. Pre-excavation estuary mud (010) with mound deposits visible beyond looking southeast. Figure 20. Pre-excavation plan of trench with (010) and [004] i n situ. Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 192 Figure 21. Northeast-facing section of slot excavated at the ed ge of the mound and estuary muds. Figure 22. Radiocarbon dates from (011) SUERC-45299 (GU30198) and (022) SUERC- 45297 (GU30196). successive expansions, augmentations, or maintenance of the monument at intervals between the 11th and 13th century A.D. (Fig. 23; O’Grady 2012:9). The remaining interior mound material was composed of re-deposited sandy clays that were bluish grey and mottled with orange, changing to orange-grey further into the mound (Figs. 23, 26). These were mixed deposits containing nodules of clay, loam patches that may be decayed turf, charcoal, and iron panning. We interpreted the color change evident across these deposits as the result of the slow leaching of brackish water through the outer bands of the mound. Radiocarbon dating and the mixed content confirmed that these were man-made deposits used to raise the core of the mound. A band of firm grey sandy clay at the base of these deposits returned a calibrated radiocarbon date, sourced from charcoal, of 1056–1281 calAD (95.4%, SUERC-45301, Betula; Fig. 27; O’Grady 2012:9). A 0.5-m sondage was excavated below this clay, revealing a thin crust of iron panning overlying a band of black/ dark brown clay with occasional charcoal fragments (possibly an old ground surface) below which was the natural substrate. This was a clean clay colored grey, turning brown-white, and comparable with a similar deposit located below the edge of the mound to the Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 193 Figure 23. Southwest-facing section of the trench excavated thr ough the mound and estuary muds. northwest, and the nearby natural clay identified by a previous excavation to the west of the site (Fig. 23; O’Grady 2012:9–10, cf. Cook 2003:4). A further calibrated radiocarbon date from above the interface of the natural substrate gave the result 1017–1207 cal AD (95.4%, SUERC-45296, Corylus; Fig. 28). This level was interpreted as the natural clay of the peninsula formed by glacial-fluvial action on the local Braemore Mudstone Formation, a mixture of mudstone, sandstone, and limestone sedimentary deposits formed approximately 398 to 407 million years ago during the Devonian Period.38 The excavation was halted at this stage (Figs. 29 and 30). Interpretation: phasing The results of the excavation can now inform an initial interpretation of the mound’s phasing, though it is of course appreciated that any model is restricted by the limitations of the evidence. It is also anticipated that what is proposed here may be refined by further paleo-environmental analysis and future larger-scale excavation. Phase I: Mid 11th–Late 12th century A.D. (Late Norse thing?/Early Gaelic Mute hill?). A natural headland, possibly earlier used as part of a thing site and located amongst a muddy inlet at the lower tidal reaches of the River Peffery estuary, was adapted by the raising of a large earth mound at some point between the latter half of the 11th century A.D. and late 12th century A.D. This phase encompassed the deposition of several layers of re-deposited clays to create the central body of the mound, the outer edges of which extended out over existing estuary mud deposits. Thin bands of clay through the center of the mound indicate where an original surface may have been. These deposits were overlain by a further thick layer which may imply that the monument was added to or adapted on at least one occasion during this period, or may have been created through a process that involved staged construction. The closely ranged sequence of radiocarbon dates, however, make it difficult to differentiate such details in the chronology. If there was not the physical evidence for phasing between the outer surfaces, these dates could in fact be used to support a single construction event for the majority of the mound. There is a small possibility that this period may have coincided with Norse incursion into Ross during the first half of the 11th century A.D. and construction of a thing mound under the aegis of the Earls of Orkney, but on balance the radiocarbon dates suggest a later event that in fact represented the re-appropriation of the thing field Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 194 and peninsula by the construction of a large court mound at the Norse meeting-place, perhaps under the orders of a resurgent earl of Ross, or a powerful mormaer of Moray such as MacBethad mac Findláich, or a representative of the Kingdom of Scots. Phase II: High medieval (Mute hill). From approximately the late 12th century A.D to the 13th century A.D., the mound was augmented with the addition of a new surface deposit or berm. This activity followed an extended period during which there was a build-up of estuary mud around the confines of the site. This phase was potentially related to the use of the site as a medieval judicial assembly mound that would eventually become associated with the development of the Scots nomenclature for the site in the place-name Mute hill, but was probably established under Gaelic lordship associated with the rulers of Ross and Moray. This phase may also be associated with increased assertion of Scottish royal power in the district, perhaps linked to a thanage and certainly with the establishment of a royal burgh. Medieval pottery and an iron vessel were deposited on the mound during this phase, perhaps simply by occasional losses associated with the nearby settlement or feasibly from encampments near the site by delegates attending judicial assemblies. From this phase the general appearance of the mound during the lifetime of the medieval burgh of Dingwall had taken shape, assuming that the lost upper form of the mound was not further augmented. Phase III: Postmedieval (mausoleum). This phase was associated with the end of the site’s function as a place of judicial assembly, and with the mound’s reuse as a burial ground for the Mackenzie family during the 17th–18th century A.D., which appropriated the site’s historic significance. Figure 24. Radiocarbon dates from the base of (011) SUERC-45300 (GU30199) and the interface of (021) SUERC-45298 (GU30197). Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 195 During this period, the rectangular enclosure wall was probably built around A.D. 1714 to surround the burial mausoleum of the 1st Earl of Cromartie. The final estuary muds were also deposited around the site during this phase, at which time midden material of animal bone and pottery was dumped into the area from the adjacent town. The channel of the tidal inlet had become largely silted-up before wholesale land reclamation followed canalization of the River Peffery between A.D. 1815 and A.D. 1817. From this point, the site became increasingly marginalized and took on the status of a local curiosity, largely owing to the memorial stone obelisk that stood on the summit of the mound. Phase IV: 20th century A.D. (car park). This was the destruction phase of the site, when in A.D. 1947 Figure 25. Northeast-facing section showing dark clay band (021 ) and clay (009). Figure 26. Part of the southwest-facing section with details of mixed clays (008), (009), and (016). Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 196 Figure 27. Radiocarbon dates from (019) SUERC-45301 (GU30200). Figure 28. Radiocarbon dates from interface of (018) with (020) SUERC-45296 (GU30195). Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 197 the mound was levelled and the car park created to serve the modern town. The memorial garden wall, still visible at the center of the car park, was also created at this time. The site had taken on its present form, and its historical significance was largely forgotten. Conclusions: archaeological investigation Although limited in scope, these archaeological investigations have confirmed for the first time the presence of important archaeological remains relating to the assembly mound, and have also Figure 30. Post-excavation view across the trench looking south east. Figure 29. Post-excavation plan of trench. Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 198 shown for the first time that this was a man-made monument created at some point between the mid- 11th century A.D. and late 12th century A.D. (Fig. 31). These are significant discoveries and make this site one of only a few purpose-built court mounds to have been excavated in the British Isles to date (see Adkins and Petchey 1984; cf. Mallett et al. 2012, O’Grady 2008b, Sanmark and Semple 2008). The Mute hill of Dingwall is also the first court mound in Scotland to have been scientifically confirmed as constructed during the medieval period (see O’Grady 2014). The identification of apparent phasing and adaptation in the mound’s construction is also a significant finding, despite that it is not possible to clarify the dating of this sequence with complete precision. We discuss the possible implications of this phasing in greater depth below. Identifying the site as a thing based purely on the archaeological remains also has difficulties. The lack of material culture pre-dating the 12th century A.D. is one factor, although absence of finds from the truncated interior of such a relatively simple monument is perhaps not entirely unexpected. The limited size of the excavation is another potential issue. Beyond the place-name evidence, which is good, several characteristics of the archaeology—such as the general lack of significant occupation deposits, the nearby and clearly separate castle mound, the close association with a field and church, and the apparent lack of medieval or earlier burials—do nevertheless provide circumstantial support for the interpretation of the site as an assembly mound, if not a thing mound. Earlier burials on the peninsula could have been destroyed by Victorian activity around the center of the site, but nevertheless such use would not preclude a later assembly function. Furthermore, an 11th-century burial mound would be extremely unusual in Scotland. Even with the complications of reasoning from negative evidence, the creation of such an extensive monument is difficult to explain in the setting of 11th–12th century Scotland other than as some form of assembly mound. This reasoning becomes particularly clear if we take into account the Norse and Scots place-name evidence and the historical context of contested Norse and Scots governance in Ross during this period, which could have influenced such a monumental statement of power and governance. The relative archaeological sterility of the mound’s clays also need not be problematic. For instance, the Secklow hundred mound, excavated in 1977–1978, was composed of simple turf and soil and lacked material culture associated with the site’s use as an assembly site (Adkins and Petchey 1984:246–250). That study also illustrated how the lack of remains can make court mounds difficult to date and how such features could be constructed into the medieval period. Another important outcome of the Dingwall excavation is that the results have discounted the notion that the site and peninsula may have been located beneath sea level in ca. A.D. 1000. Further paleo-environmental research into historic ocean levels along the Cromarty Firth coast and the extent of glacial rebound will help to refine this finding. More information about the site’s environs and construction will also be forthcoming from paleoenvironmental analysis of bulk soil samples and thin-section samples taken during the excavation Figure 31. Calibrated radiocarbon dates from the Mute hill. Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 199 that are to be published elsewhere. Nonetheless, it is clear from these initial findings that a substantial amount of man-power and effort would have been required to create a monument on the scale of the Dingwall Mute hill, at 38 m by 48 m, which in plan is around half the size of the royal mound at Scone. In simplified terms, the large size of the mound, and the substantial marshalling of manpower required to create it, would seem in keeping with the establishment of a major regional center for legal organization. But can we reconstruct further the political and historical context for creation of this extraordinary monument? The final part of the article will consider these questions based on our reasoned understanding of the site’s development and historic landscape with a new review of recent research into thing sites found elsewhere in Scotland. Scottish Thing Sites Since Fellows-Jensen’s (1993, 1996) pivotal survey, six new thing place-names have been identified in Scotland. All of these are in the Western Islands of Scotland, on the Hebridean islands of Bute, Islay, Mull, Eigg, and Lewis. Edin on Bute (Atyng’ar 1475, Iding 1577) is possibly derived from ON althing (“a regional or head assembly”) or ON eið and ON þing “isthmus-assembly”, and associated with an enclosed mound, Cnoc an Rath in Borgadale. This name contains ON borg “fort, dome-shaped hill” and dalr “valley” (Duffy 2011; Markus 2012a:8, 2012b; O’Grady 2011b). Sunderland on Islay in Kilchoman parish is possibly from ON *Sjóvarþing, “the assembly place by the lake”, which Macniven identified with a prominent and flat-topped natural mound overlooking Loch Gorm (Caldwell 2008:333; Macniven 2006:363, 364, fig. 74, 365). Whyte has recently refuted Macniven, opting instead for a cairn at nearby Grulin, ON grjót and ON þing “stony thing”. Whyte has suggested a similar interpretation for the place-name Gruline on the Isle of Mull that is located near a large flat-topped stony mound, and also for Grulin on the Isle of Eigg (Whyte 2014:117–119, 139–147). In the Outer Hebrides, Eilean Thinngartsaigh is a small and uninhabited rocky island off the south coast of Lewis in Loch Claidh39 and is thought to contain a thing element (B. Crawford, School of History, Univesity of St. Andrews, Fife, Scotland, UK, pers. comm.). Also on Lewis is Toingal that can be linked to a small hill in Uig parish (Cox 1991:484). More tentatively, Maeshowe chambered cairn on Orkney has been proposed as a thing due to the unusual concentration of runic inscriptions at the site and the heightening of the surrounding bank and central mound during the Viking Age. Kirkwall, also on Orkney, has been identified as a thing, though mostly based on saga evidence (Graham-Campbell and Batey 1998:42, 61). This small group of new sites provide a valuable expansion on the 11 examples established by Fellows-Jensen and previously known from the Shetland Islands (Jakobsen 1993, cf. Sanmark 2013). These additions bring the total to 22 thing place-names from Scotland (Fig. 32). The landscape form of these sites are not all the same, which as we shall see, is probably also true of their historical development. In addition to the aforementioned group, Owen and Driscoll (2011) have recently repeated a case for Norse influence in the development of an assembly mound at Govan in central Scotland, the lost “Doomster Hill”. Over the last two decades, several published articles have assigned Govan a prominent position in the established literature on Scottish assembly sites and Norse influence therein (see Driscoll 1998, Owen and Driscoll 2011:345). This prominence in itself would require us to consider Govan in more detail, though there is no Norse place-name at Govan40 and the archaeology of the Doomster Hill’s remains largely unknown, but some of the interpretive reasoning used to link Govan to Norse influence has particular relevance for our discussion of Dingwall. This is perhaps not the place for an extended critique of Owen and Driscoll, but it is nevertheless worthwhile to briefly review the main strands of evidence put forward for Govan’s Norse association. The Doomster Hill does not survive, and as such, former discussion has focused on associated circumstantial evidence at the nearby church and comparison with other sites, in particular Tynwald Hill on the Isle of Man (Owen and Driscoll 2011:342). A radiocarbon date spanning the early 8th to late 9th century A.D. was derived from within a surfaced road at the historic entrance to the nearby churchyard. This finding was interpreted as evidence for a “ceremonial route” between the church and mound, and was suggested as comparable with arrangements at the Tynwald Hill and St. John’s church on the Isle of Man (ibid:340–341). Deposits indicating industrial activity were also uncovered. The burial ground, mound and a purported royal residence on the opposite bank of the Clyde were proposed as part of a “power center”; this model was alluded to again in a recent article with little further qualification (Driscoll 2016:74, 79, 89). The Doomster Hill was also argued to have a ditch and stepped flanks. This was based on interpretation of an 18th-century A.D. illustration of the site, and again compared with Tynwald Hill and 3 other possible stepped-mounds in the Irish sea zone at Dublin, Cumbria, and perhaps Dumfriesshire (Darvill Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 200 2004:229, fig. 10.6; O’Grady 2008a:211–217; Owen and Driscoll 2011:341, fig. 7).41 A narrative for the site’s Norse affiliation was also proposed around the record of a raid by Dublin Vikings in 870 A.D. on Dumbarton Rock, a major fortress of the Northern Britons on the River Clyde estuary.42 The associative evidence of five hogback grave-markers located at Govan’s old parish church has also been crucial to the Doomster Hill case. Linked on stylistic grounds to northern England, the Govan hogback monuments form part of a larger collection at the site of 10thcentury cross-slabs and free-standing crosses. The cultural implications of the hogback monuments’ artistic connections have recently undergone significant revision and remain in contention (O’Grady 2003, Ritchie 2004, Williams 2015; cf. Driscoll et al. 2005). Recent research into assembly sites and early medieval church settlements elsewhere in Scotland can support an alternative view, which we argue raise some complications for Owen and Driscoll’s (2011:343) assertion that “a high status Norse-style thing place” existed at Govan. We would not refute that wider comparisons with sites in other parts of the British Isles and in Scandinavia is worthwhile, as we have suggested for Dingwall, but we do contest that Govan should foremost be understood in relation to patterns of early medieval archaeology in Scotland. For instance, new research has shown that the pairing of churches and assembly mounds was a particularly common aspect of early medieval assembly mounds and hills in Scotland (O’Grady 2014). Prominent examples include the Bishop’s Hill by Dunkeld Cathedral, Tillydrone Hill by St. Machar’s Cathedral in Aberdeen, Mortlach in Banffshire, and Moothill located beside the Abbey of Scone (ibid:114, 116–117, 119–122, 123–125). Such links between church and assembly sites were known to a lesser extent in parts of Scandinavia and Tynwald on the Isle of Man (Brink 2003, Darvill 2004). Sanmark (2013:104) has also noted that a key feature of thing sites in Shetland is closeness to churches, though some of these chapels may be medieval, and the exact relationship between the sites has not been verified. The larger group of churches and assembly sites are, however, mostly evidenced in mainland Scotland and suggest a separate development in northern Britain. The Dingwall and Govan assembly mounds cannot therefore be solely considered as developing under Norse hegemony on the basis of proximity to a church, but alternatively may be related to developments elsewhere in Scotland in areas outside of Norse political dominance. Secondly, although it was asserted that there is “no evidence at all” for a monastery at Govan (Owen and Driscoll 2011:334), the significant concentration Figure 32. Map of thing sites in Scotland. (Background map reproduced from Ordnance Survey map data by permission of Ordnance Survey © Crown copyright 2016.) Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 201 of sculpture at the church is a feature almost exclusively seen in Scotland at important early medieval monasteries and major church settlements, such as Iona (RCAHMS 1982), St. Andrews (RCAHMS 1933:237–239), Portmahomack (Carver 2008), and Fortingall (O’Grady 2013, Robertson 1997), among others, an exception being the small burial ground at Cladh a’Bhile, Argyll (Gondek 2006). Moreover the metalled roadway or “ceremonial route” uncovered at Govan church has its only comparators in Scotland in recent excavations at church settlements, such as the kerbed roadway at Portmahomack, the metalled entrance road at Fortingall, and the wellknown “road of the dead” at Iona (Carver 2008, O’Grady 2013, RCAHMS 1982). The radiocarbon date for the Govan road is also open to reinterpretation. The date range overlaps the 870 A.D. raid by around only 20 years and it may be significant that the sample was derived from a deposit that the excavator described as a “repair” in the road (Owen and Driscoll 2011:340–341). It would seem reasonable to conclude that the road was in use prior to recorded Viking incursion into Strathclyde. Could this route way simply have formed part of the established layout and ceremonial space of a larger church settlement? Though this would be late for a flourishing early medieval monastic settlement in Scotland, this revision would not preclude the existence of a larger royal church-settlement at Govan, comparable to the klosterpfalzen or “monastic palace” of Germany and Late Anglo-Saxon England (cf. Woolf 2010:313), and not out of place among other 10th–11th century Scottish royal centers such as Dunfermline and Scone. The arrangement of church and mound is in keeping with the wider pattern of Scottish judicial assembly sites (see O’Grady 2014), which is unlikely to have wholly derived from Norse influences and implies the close role of Church authorities in the development of early law in Scotland. This interpretation would not require a Norse explanation for the creation of the Doomster Hill. A more nuanced reading of the evidence seems advisable, in particular given the interplay of British, Norse, and early Gaelic influences on late 1st millennium AD Govan (see Macquarrie 1994). The mutability and propensity for reuse and adaptation of assembly sites is increasingly understood as an important process (O’Grady 2014, Sanmark and Semple 2008). Assembly sites could be redeveloped and adapted and enhanced relatively late into the 11th century, as we have seen in the excavated evidence for the Dingwall mound, but they could also incorporate earlier places as at Tingwall, Shetland (see above). Such factors of reuse and appropriation could be relevant for understanding the Doomster Hill and potential for Norse assimilation of the site in the late 9th century A.D., but at present it is not possible to confirm the extent that these factors play a role there. It is essential that theories exploring the development of assembly sites are elucidated with excavation data, particularly because it is apparent that assembly sites can have unusually complex histories. For instance, though admittedly the evidence is limited, a stricter reading of the post-medieval records and place-names seems to indicate that the Doomster Hill may have been some form of reused burial mound, perhaps prehistoric (Davidson-Kelly 1994; NSA 1834–1845:v.6, 690; OSA 1791–1799:v.14, 294). In the absence of further excavations, this must remain one of several possibilities, but this material should not be wholly discounted and reminds us of the potential complexity of site biographies. The name “Doomster” indicates that the mound was used as a court hill during the late medieval period, as is also suggested with the Dingwall Mute hill. Driscoll (2004) has asserted that Viking traditions may have affected the wider development of assembly mounds in Scotland, even at important royal sites such as the Moothill at Scone. This conclusion, however, seems premature as the current excavated archaeological evidence provides little support for this idea. Instead, the implications from landscape and placename evidence suggest that Govan formed part of a common grouping of early medieval churches and assembly mounds found elsewhere in lowland and northeast Scotland. On this basis, Driscoll’s view of incoming Norse traditions is also at odds with the wider evidence for parallel traditions of assembly mounds and hills found across northern Europe, including in particular, though not exclusively, Anglo- Saxon England, Ireland, Scandinavia, and Scotland (Gleeson 2016 [this volume]; O’Grady 2008a, 2014; Reynolds 2013; Sanmark and Semple 2008). The question of whether these traditions developed from earlier expressions of power and monumentality centered around burial mounds or required external influences needs further urgent research (see Fitz- Parick 2004, Semple 2004). In order to adequately advance assembly studies in Scotland, there is a clear imperative for future research to source firm archaeological data from clearly documented sites, which can then be used to inform more meaningful broader interpretation of this site-type. Such efforts will help to reduce our current reliance on comparisons with other traditions and better known sites from outside Scotland. An admirable recent example of a conscientious scientific study is Coolen and Mehler’s (2014) explorations at Tingwall, Shetland. Their study has illustrated the potential complexities of landscape Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 202 reuse at North Atlantic thing sites. Our explorations at Dingwall also illustrate that it is possible to derive meaningful archaeological information from sites that are located in modern urban landscapes substantially affected by recent development. These two key studies also highlight the importance of methodologies that incorporate detailed local historical research with well-planned campaigns of geophysical surveys prior to attempting excavation at assembly sites (O’Grady 2011a, 2012). Future GPR survey work at Govan could still yield more useful results. Govan aside, our focus here is on the types of locations and sites associated with thing place-names in Scotland, as these will have important implications for interpretation of the Dingwall mound. No full consideration of the group’s archaeology has so far been undertaken (cf. Crawford 1987:204–210, Whyte 2014). Where sites can be identified, mounds are the most common and include 8 possible thing mounds, not counting the Dingwall Mute hill (Fig. 32). However, a closer look reveals important distinctions in the different types of mounds that make up the group. At least two internal types can be isolated. The first are prehistoric mounds, either settlement mounds or burial monuments, which include 3 sites focused around Orkney and the adjacent Caithness coast or 5 if Maeshowe and Grulin on Islay are included.43 These sites appear to include turfcovered remains of Iron Age brochs, large dry-stone built towers, which pre-dated the Viking Age. The only excavation in the wider group at Tingwall on Shetland has also uncovered evidence for Iron Age occupation at the island (Coolen and Mehler 2014), but most of the dating of these mounds is admittedly based only on interpretation of the surface remains.44 Nevertheless, these prehistoric mounds imply a dependency on the appropriation of existing landscape features for the siting of thing assemblies, as may have been the case at Govan, and presumably during the settlement of the Northern Isles from the 9th century A.D. onwards. In this sense, the two natural hill sites can also be included with this group.45 There are also indications of possible site adaptation, such as at Thing’s Va in Caithness, where the broch-mound is associated with a possible enclosure and nearby turf-covered cairn, and at Maeshowe as mentioned above.46 Such sites may also indicate the selection of landscape features by Viking elites who still adhered to pagan cult beliefs and legal observance, in which pre-Christian ideas about the association of burial mounds with supernatural powers and royal ancestry may have played an important role (see Ellis 1977). The second group cannot be so well defined— again due their poor survival and inconsistent topography—but in general terms these comprise simpler mounds, which in some cases may have been created for the purposes of assembly during the Viking Age.47 The ambiguous earthwork called the Cnoc an Rath on the Isle of Bute, which is possibly linked to the thing place-name Edin, was set within a landscape of prehistoric monuments and apparently surrounded by a ditch and bank. Recent excavation at the site initially proved inconclusive due to post-medieval disturbance. However, recent post-excavation analysis has provided 2 radiocarbon dates from a preserved surface at the site that returned calibrated dates of the 7th to 9th century A.D., which may suggest Viking Age activity at the site associated with assemblies (P. Duffy, Brandanii Archaeology, Rothesay, Isle of Bute, UK, pers. comm.; O’Grady 2011b). At Tinwald in Dumfriesshire, geophysical survey has provided a preliminary indication that the candidate mound may have had a stepped profile, but this detail remains unclear due to the site’s use as a later motte castle during the high medieval period. An alternative site for the thing is Tinwald Downs, a level plain more in keeping with the völlr “field” place-name, where military musters and horse races were held up to the 19th century A.D. (O’Grady 2008a:211–216). The flat-topped cairn or mound at Gruline on Mull may also be relevant here (Whyte 2014:117–119). Again obtaining a coherent archaeological data set for the group is a key priority for future research. As the only purpose-built mound so far verified in Scotland, it is difficult to set the Dingwall mound in context with other sites in the Atlantic north. Nevertheless, the propensity for mounds and mound-like hills at other thing sites that were linked to þing-völlr place-names could provide circumstantial comparative basis on which to accept the Dingwall Mute hill as a reused thing mound.48 On the other hand, mounds and particularly hills are also very common amongst assembly sites linked to early Gaelic or Scots place-names in other parts of Scotland, and throughout England and wider northern Europe (O’Grady 2014, Reynolds 2013, Sanmark and Semple 2013). The mound site-type can therefore not be used alone to argue that the Dingwall mound was created by Norsemen. This is also true of the association of assembly mounds with early church sites, because, as discussed above, this is a feature found in Scotland and to some extent also in Scandinavia (Brink 2003, O’Grady 2014). Whether this is a shared feature of legal cultures in the Northern World during the Viking Age or reflects pan- European influence by the Church in early medieval judicial organization beyond royal centers is not as yet fully understood (see O’Grady 2014:113–117, 119–123, 131). Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 203 Dingwall’s landscape can also be compared more widely with other thing sites. The Mute hill’s location at the head of a peninsula with direct access to water route-ways and a close-by freshwater supply within a centrally located and fertile valley are also landscape attributes found at the Shetland lawting in the parish of Tingwall. Recent excavation at the island has revealed a natural feature previously used for late prehistoric settlement (Coolen and Mehler 2014). Though admittedly the Law Ting Holm lacks direct coastal access, this shortcoming is made up for by the site’s centrality within the Shetland archipelago, and the general similarities in landscape characteristics, nonetheless, remain striking with Dingwall. Even more closely matched with Dingwall’s landscape is the important thing site at Nidaros (Trondheim) on the Trondheimsfjord in Norway that was known as the Eyraþing. At Nidaros, the thing was located on a strand at the mouth of the River Nid. This location was close by the trading center and settlement established in A.D. 997 on a wide bend in the river from which the settlement takes its name. Adjacent is St. Olaf’s cathedral founded in A.D. 1031, with its palace complex and medieval inauguration site associated with the cult of St. Olaf (Øystein 2014; E. Øystein, Nidaros Cathedral Restoration Workshop, Trondheim, Norway, pers. comm.). Similarly the Gulatinget thing, established in the early 10th century A.D., was sited on a field and natural outcrop at an accessible coastal location near the mouth of the Sognefjord in southwest Norway (Brink 2003). These key regional thing sites from the Norse homeland provide useful comparative exemplars from which to interpret the proposed thing complex at Dingwall with its field and church. It is notable though that none of these three examples of important things included an assembly mound. Was the Dingwall thing sited based on a model with these pre-eminent and regionally important sites in mind? If so was the assembly mound added subsequently and was this done under the aegis of Norse or Scottish authorities? Perhaps by the mid-11th century A.D., such a difference could be indistinguishable in the archaeological record because of the close similarities between traditions of legal assembly on mounds in northwest and northeast Scotland? The continued use of Tingwall in Shetland into the 16th century A.D. also provides a context within which to understand the apparent extended use of the Mute hill at Dingwall into the medieval period. Athough in the case of the district of Ross this usage would have been within the jurisdiction of Scots Common Law rather than the authority of the Kingdom of Norway as in the north. Such longevity into the late medieval period was also a feature of judicial assembly sites more widely in the Kingdom of Scots (O’Grady 2014:119, 126, 131). We also need to factor in sufficient time to have elapsed prior to the 13th century A.D. for the active thing place-name to have stabilized at Dingwall. On balance, it would seem more likely that the thing was already established at Dingwall prior to the creation of the mound from the 11th century A.D. The parallel Scandinavian and Scottish traditions of assembly mounds therefore leave us with an interpretive choice. The construction of the mound can thus either be seen as a final attempt to assert Norse authority, by creating a more impressive monumental statement at the site or, perhaps more likely, a reassertion of authority by the lords of Ross who were drawing on an established motif of Gaelic legal organization and lordship, and which potentially was also meant to reference the “royal” Moothill at Scone. As a means of concluding the article, we will now draw together our findings to reflect on the most likely historical context for the creation of the Dingwall Mute hill and the implications this has for understanding Scottish thing sites more generally. Conclusion: A Late Norse or Early Gaelic Assembly Mound? This article has set out to achieve a more accurate understanding of the thing site at Dingwall, which had previously only been known from place-name evidence. Our intentions were to better understand the precise location of the assembly site, its historic landscape, and archaeological remains. In the absence of direct historic evidence, a lead candidate has been proposed in the mound known in medieval records as the Mute hill of Dingwall. Based on comparative analysis of the landscape with other thing sites, thorough reconstruction of the site’s topography by detailed documentary research, and archaeological field investigations, we have made a case for creation of the site as an assembly mound around the mid-11th century A.D. How though should this construction date be understood in the political context of Late Norse Period northern Scotland? We have reasonably proposed several possible scenarios, though the limitation in our evidence means that no one argument can be put forward as the definitive explanation, particularly in the absence of Norse material culture from the site. On the one hand, if we accept that at least some form of large-scale and politically motivated military assault, or assaults, were made on Caithness and Ross by Norse earls during the first half of the 11th century A.D., then this would seem an attractive Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 204 context in which to explain the creation of a new and impressive legal administrative site such as the Mute hill mound. There are strategic and geopolitical characteristics of the site that also make this a plausible suggestion. Dingwall and the Cromarty Firth mark a key border zone between Morayshire in the south and the important nexus of routeways centered on Strathpeffer that provide access between the west coast of Scotland and the North Sea. This border-zone concept was also supported by place-name evidence for the southerly extent of Norse settlements on the east coast, which falls on the Cromarty Firth–Black Isle line. Moreover, the notion of Christian Late Norse earls establishing a power base at Dingwall is circumstantially supported by the presence of a church foundation beside the mound site linked to the cult of St. Clement, a cult which has strong connections to early Scandinavian Christian observance in the 11th century A.D. (see Crawford 2008). The establishment of a new Norse church foundation at Dingwall could also be seen as an attempt to usurp the traditional ecclesiastical center of the region at Rosemarkie, located at the eastern tip of the Black Isle, 38 km east of Dingwall (Woolf 2006; cf. Crawford 1995), and compliments the idea of the creation of a new power center. Finally the close similarities of Dingwall’s landscape characteristics to other important thing sites in the Norwegian homeland and in the Shetland Islands suggest that the site may have been created with these existing models in mind by protagonists who were familiar with such sites. However, as noted above, these comparators lack assembly mounds. On the other hand, the landscape characteristics and siting on a coastal peninsula are features rarely found in combination at early medieval assembly sites further south in mainland Scotland (O’Grady 2014). Based on these factors, a Norse agency can be argued for convening of the thing in the 11th century A.D, and possibly associated with the traditions of incursions made by the Earls of Orkney into northern Scotland during this period, but not necessarily in the creation of the mound. This interpretaion would of course not preclude the possibility that a thing may have been first sited at the headland or nearby field site as part of Viking conquest in the area during the 9th century A.D. and before the creation of the mound. Given the doubt cast over the saga evidence, it would be imprudent to put forward a named candidate for the individual behind the creation or re-establishment of the thing, even if we accept the above scenario and despite the obvious temptation to bestow this honor on Thorfinn the Mighty of Orkney. Indeed it is the nuanced interaction of Late Norse and Scottish interests at Dingwall that can inform a more compelling interpretation of the assembly mound. It is clear that by the first decades of the 12th century A.D. the region of Ross and presumably Dingwall were under the lordship of mormaers of Ross and not under Norse control. This chronology would seem to imply that either the use of the mound for a thing was only short-lived as a focus for Norse legal organization, perhaps only for a generation, or that Norse annexation of the area was not sustained during the 11th-century A.D.. The former might explain the apparent lack of a developed system of ounceland divisions in the area. If the latter, then it might be possible that the assembly mound was in fact created by a mormaer of Ross to reassert judicial authority in the region, drawing on contemporary early Gaelic assembly practices within the Kingdom of Alba (see O’Grady 2014). The phased construction of the mound between at least the 11th and 13th centuries A.D. could also be associated with augmentation of the site following reassertion of Scottish royal authority in the area. Such an event might also fit into the much-debated attempts made by the mormaers of Moray to assert their sovereignty during the 11th century A.D. The implication of this interpretation is that we accept that the thing had been an existing institution which pre-dated the mound’s creation, perhaps either sited on the natural peninsula where the mound was later created or at the nearby level plain of the Trinity field. If we favor this model, then the possible period for establishment of the thing may have followed the devastation of Fortriu in the 9th century A.D. The radiocarbon dates from the mound certainly do not sustain with any certainty a pre-11th century construction date, and although the physical evidence for phasing is good, the date-span for the 6 radiocarbon dates between the mid-11th to late-13th century A.D. makes more detailed interpretation inadvisable. There is a clear need for more excavated examples of assembly sites in Scotland to compare with the currently singular results from the Dingwall mound, and future studies should seek to investigate and recover dating evidence from a larger sample of assembly mounds. Future GPR survey and excavation in the area of Trinity Field may also reveal evidence for the thing. Nonetheless, with this article, we have made a significant contribution by presenting the first archaeological evidence for a purpose-made assembly mound in Scotland. We have substantially increased understanding of the structural and chronological development of the Mute hill of Dingwall, and highlighted historic landscape features important for the study of thing sites in Scotland. Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 205 Acknowledgments This article is based on research by the authors for the Dingwall Thing Project, which was supported by funding grants from the Society of Antiquaries of Scotland, Hunter Archaeological and Historical Trust, and Strathmartine Trust. The authors wish to express their grateful thanks for funding received from the THING Project (funded by the EU Northern Periphery Programme 2007–2013), The Highland Council and Dingwall History Society in support of archaeological excavation at the Cromartie Memorial Car Park in Dingwall. We are also grateful to The Highland Council for giving access permission to the car park for the purposes of fieldwork. The thinking which underpins the ideas presented in this article was developed independently by Oliver O’Grady during research for his doctoral thesis and by David and Sandra MacDonald during local research for the Dingwall History Society. The resulting article is the outcome of the authors’ subsequent collaborative research. Sincere thanks to Alexandra Sanmark for the invitation to contribute to this volume and for the diligent work of the editor and anonymous reviewers. Any errors within remain the authors. This article is dedicated to the late David MacDonald, who passed away during the editorial process. Abbreviations NSA 1845. The new statistical account of Scotland: By the ministers of the respective parishes, under the superintendence of a committee of the society for the benefit of the sons and daughters of the clergy. W. Blackwood and Sons, Edinburgh, UK. OSA Sinclair, J. (Ed.) 1791–1799. The [old] statistical account of Scotland: Drawn up from the communications of the ministers of the different parishes. W. Creech, Edinburgh, UK. Literature Cited Adkins, R.A., and M.R. Petchey. 1984. Secklow hundred mound and other meeting-place mounds in England. Archaeological Journal 141:243–251. Anderson, M.O. 1973. Kings and Kingship in Early Scotland. Scottish Academic Press, Edinburgh, UK. 310 pp. Bain, R. 1899. History of the Ancient Province of Ross. The Pefferside Press, Dingwall, UK. 480 pp. Baker, J., and S. Brookes. 2015. Identifying outdoor assembly sites in early medieval England. Journal of Field Archaeology 40.1:3–21. Barrow, G.W.S. 1960. Regesta Regum Scottorum Vol. 1. Edinburgh University Press, Edinburgh, UK. 339 pp. Barrow, G.W.S. 1992. Scotland and its Neighbours in the Middle Ages. Hambledon Press, London, UK. 255 pp. Benediktsson, J. 1986. Íslendingabók, Landnámabók. Hiđ Íslenzka Fornritafélag, Reykjavík, Iceland. 525 pp. Brink, S. 2003. Legal assemblies and judicial structure in early Scandinavia. Pp. 61–72, In P.S. Barnwell and M. Mostert (Eds.). Political Assemblies in the Earlier Middle Ages. Studies in the Early Middle Ages 7. Brepols, Turnhout, Belgium. 213 pp. Brink, S. 2004. Legal assembly sites in early Scandinavia. Pp. 205–216, In A.Pantos and S. Semple (Eds.). Assembly Places and Practices in Medieval Europe. Four Courts Press, Dublin, Ireland. 256 pp. Caldwell, D. 2008. Islay the Land of Lordship. Birlinn, Edinbugh, UK. 320 pp. Carver, M.O.H. 2008. Portmahomack: Monastery of the Picts. Edinburgh University Press, Edinburgh, UK. 240 pp. Cavill, P., S. Harding, and J. Jesch. 2004. Revisiting Dingesmere. Journal of the English place name society 26(October 2004):25–38. Clancy, T.O. 1996. Govan: The name. Annual report of the society of friends of Govan old 6:2–3. Clark, A. 1993. Vikings in Ross and Cromarty: Fact and Fantasy. Museums Section of Ross and Cromarty District Council Leisure Services, Dingwall, UK. Cook, M. 2003. Tesco, Dingwall Action Mart site: Archaeological works data structure report. AOC Archaeology Group. Unpublished report. Available online at http://her.highland.gov.uk/hbsmrgatewayhighland/ DataFiles/LibraryLinkFiles/9581.pdf. 11 pp. Coolen, J., and N. Mehler. 2014. Excavations and Surveys at the Law Ting Holm, Tingwall, Shetland. An Iron Age Settlement and Medieval Assembly Site. BAR British Series 592. Archaeopress, Oford, UK. 135 pp. Cowan, E.J. 1993. The historical Macbeth. Pp. 117–142, In W.D.H. Sellar (Ed.). Moray: Province and People. Scottish Society for Northern Studies, Edinburgh, UK. 264 pp. Cowper, H.S. 1891. Law ting at Fell Foot, Little Langdale, Westmorland. Transactions of the Cumberland and Westmorland Antiquarian and Archaeological Society 11:1–6. Cox, R.A.V. 1991. Norse-Gaelic contact in the west of Lewis: The place-name evidence. Pp.479–493, In P. Sture Ureland and G. Broderick (Eds.). Language Contact in the British Isles. Proceedings of the Eighth International Symposium on Language Contact in Europe, Douglas, Isle of Mann, 1988. M. Niemeyer, Tübingen, Germany. 717 pp. Crawford, B.E. 1986. The making of a frontier: The Firthlands from 9th–12th centuries. Pp. 33–46, In R.J. Baldwin (Ed.). Firthlands of Ross and Sutherland. Scottish Society for Northern Studies. Edinburgh, UK. 220 pp. Crawford, B.E. 1987. Scandinavian Scotland. Leicester University Press, Leicester, UK. 280 pp. Crawford, B.E. 1995. Earl and Mormaer: Norse-Pictish Relationships in Northern Scotland. Groam House, Rosemarkie, UK. 30 pp. Crawford, B.E. 2008. The Churches Dedicated to St. Clement in Medieval England. A Hagio-geography of the Seafarer’s Saint in 11th-century North Europe. Scripta Ecclesiastica 1. Axioma, St. Petersburg, Russia. 237 pp. Crawford, B.E., and S. Taylor. 2003. The southern frontier of Norse settlements in North Scotland place-names and history. Northern Scotland 23:1–76. Curle, A.O., M. Olsen, and H. Shetelig. 1954. Viking Antiquities in Great Britain and Ireland. Civilisation of the Viking Settlers in Relation to Their Old and New Countries 6. Aschehoug, Oslo, Norway. 251 pp. Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 206 Dalrymple, D. 1770. The Additional Case of Elisabeth, Claiming the Title and Dignity of Countess of Sutherland, by her guardians. Unknown publisher, London, UK. 177 pp. Darvill, T. 2004. Tynwald Hill and the “Things” of power. Pp. 217–233, In A. Pantos, and S. Semple (Eds.). Assembly Places and Practices in Medieval Europe. Four Courts Press, Dublin, Ireland. 256 pp. Davidson-Kelly, T.A. 1994. The Govan collection in the context of local history. Pp. 1–19. In A. Ritchie (Ed.). Govan and its Early Medieval Sculpture. Alan Sutton Publishing Limited, Stroud, UK. 168 pp. Driscoll, S.T. 1998. Church archaeology in Glasgow and the kingdom of Strathclyde. The Innes Review 49.2:95–114. Driscoll, S.T. 2004. The archaeological context of assembly in early medieval Scotland: Scone and its comparanda. Pp. 73–94, In Pantos, A., and S. Semple (Eds.). Assembly Places and Practices in Medieval Europe. Four Courts Press, Dublin, Ireland. 256 pp. Driscoll, S.T.D. 2016. Reading Govan Old: Interpretative challenges and aspirations. Pp. 73–92, In F. Hunter and A. Sheridan (Eds.). Ancient Lives: Objects, People, and Place in Early Scotland. Essays for David V Clarke on his 70th birthday. Sidestone Press, Leiden, The Netherlands. 382 pp. Driscoll, S.T., O.J.T. O’Grady, and K. Forsyth. 2005. The Govan school revisited: Searching for meaning in the early medieval sculpture of Strathclyde. Pp. 135–158, In S. Foster and M. Cross (Eds.). Able Minds and Practised Hands. Society for Medieval Archaeology, Leeds, UK. 434 pp. Duffy, P. 2011. Cnoc an Rath, Argyll, and Bute (North Bute parish): geophysical survey and excavation. Discovery and Excavation in Scotland New 12:52–53. Dumville, D.N. 1997. The Churches of North Britain in the First Viking Age. Friends of the Whithorn Trust, Whithorn, UK. 38 pp. Ellis, H.R. 1977. The Road to Hel: A Study of the Conception of the Dead in Old Norse Literature. Reprint of 1943 edition. Greenwood Press, Westport, CT, USA. 208 pp. Fellows-Jensen, G. 1993. Tingwall, Dingwall, and Thingwall. North-Western European Language Evolution 21/22:53–67. Fellows-Jensen, G. 1996. Tingwall: The significance of the name. Pp. 16–29, In D.J. Waugh (Ed.). Shetland’s Northern Links: Language and History. Scottish Society for Northern Studies, Edinburgh, UK. 256 pp. FitzPatrick, E. 2004. Royal Inauguration in Gaelic Ireland c. 1100–1600. A Cultural Landscape Study. Studies in Celtic History XXII. Boydell, Woodbridge, UK. 294 pp. Fraser, I. 1986. Norse and Celtic place-names around the Dornoch Firth. Pp. 23–32, In R.J. Baldwin (Ed.). Firthlands of Ross and Sutherland. Scottish Society for Northern Studies, Edinburgh, UK. 220 pp. Fraser, W. 1876. The Earls of Cromartie: Their Kindred, Country, and Correspondence. I/II. Thomas and Archibald Constable, Edinburgh, UK. 720 pp. Gleeson, P. 2015. Kingdoms, communities, and Óenaig: Irish assembly practices in their Northwest European context. Journal of the North Atlantic Special Volume 8:33–51. Gondek, M. 2006. Early Historic Sculpture and Landscape: A case study of Cladh a’Bhile, Ellary, Mid- Argyll. Proceedings of the Society of Antiquaries of Scotland 136(2006):237–258. Graham-Campbell, J. 2004. “Danes … in this country”: Discovering the Vikings in Scotland. Proceedings of the Society of Antiquaries of Scotland 134(2004):201–239. Graham-Campbell, J., and C.E., Batey. 1998. Vikings in Scotland: An Archaeological Survey. Edinburgh University Press, Edinburgh, UK. 296 pp. Grant, A. 2000. The province of Ross and the kingdom of Alba. Pp. 88–126, In E.J. Cowan and R.A. McDonald (Eds.). Alba: Celtic Scotland in the Middle Ages. Tuckwell Press, East Linton, UK. 282 pp. Hudson, B.T. 1998. The Scottish chronicle. Scottish Historical Review 77:129–161. Hunter, F. 2011. Pitgrudy, Highland (Dornoch parish), metal detector find. Discovery and Excavation in Scotland, New 12:104. Innes, C. 1840. Liber Cartarum Sancte Crucis. Bannatyne Club, Edinburgh, UK. 333 pp. Innes, C. 1872. Lectures on Scotch Legal Antiquities. Edmonston and Douglas, Edinburgh, UK. 326 pp. Jakobsen, J. 1993. The Place-names of Shetland. Facsimile of 1936 edition. Orcadian, Kirkwall, UK. 273 pp. Johnston, J.B. 1934. Place-names of Scotland. 3rd Edition. John Murray, London, UK. 335 pp. Lawrie, A.C. (Ed.). 1905. Early Scottish Charters Prior to 1153. MacLehose, Glasgow, UK. 515 pp. Macniven, A. 2006. The Norse in Islay. Ph.D. Dissertation. University of Edinburgh, Edinburgh, UK. 550 pp. Macquarrie, A. 1994. The historical context of the Govan stones. Pp. 27–32, In A. Ritchie (Ed.). Govan and its early medieval sculpture. Alan Sutton Publishing Ltd., Stroud, UK. 168 pp. Macrae, N. 1923. The Romance of a Royal Burgh: Dingwall’s Story of a Thousand Years. North Star Proprietors, Dingwall, UK. 391 pp. Mallett, L., S. Reddish, J. Baker, S. Brookes, and A. Gaunt. 2012. Community archaeology at Thynghowe, Birklands, Sherwood Forest. Transactions of the Thoroton Society of Nottinghamshire 116:53–72. Markus, G. 2012a. From goill to gall-ghàidheil: Scandinavian settlement in Bute. Pp. 1–16, In Ritchie, A. (Ed.). Historic Bute: Land and People. Scottish Society for Northern Studies, Edinburgh, UK. 162 pp. Markus, G. 2012b. The Place-Names of Bute. Shaun Tyas, Donington, UK. 624 pp. McDonald, R.A. 2003. Outlaws of Medieval Scotland, Challenges to the Canmore Kings, 1058–1266. Tuckwell, East Linton, UK. 202 pp. Miller, H., and D. MacLeod. 1889. Notice of the discovery of a hoard of silver penannular armlets and coins at Tarbat, Ross-shire. Proceedings of the Society of Antiquaries of Scotland 23:314–317. Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 207 O’Grady, O.J.T. 2003. An interdisciplinary re-evaluation of the early medieval sculptured crosses of the Glasgow area. M.Phil. Dissertation. University of Glasgow, Glasgow, UK. 90 pp. O’Grady, O.J.T. 2008a. The setting and practice of openair judicial assemblies in medieval Scotland: A multidisciplinary study. Ph.D. Dissertation. University of Glasgow, Glasgow, UK. 649 pp. O’Grady, O.J.T. 2008b. Tracing the medieval royal centre at Scone. Medieval Archaeology 52:376–378. O’Grady, O.J.T. 2011a. Dingwall viking thing project: Cromartie memorial car park, Dingwall, Highland. Report on geophysical survey: Ground penetrating radar. Unpublished report. Available online at https:// glasgow.academia.edu/OliverOGrady. 33 pp. O’Grady, O.J.T. 2011b. Cnoc an Rath, St Colmac’s, Isle of Bute,. Report on geophysical survey: Electrical resistance. Unpublished report. Available online at https:// glasgow.academia.edu/OliverOGrady. 23 pp. O’Grady, O.J.T. 2012. THING project: Cromartie memorial car park, Dingwall, Highland. Archaeological excavation: Data structure report. Unpublished report. Avialable online at http://her.highland.gov. uk/hbsmrgatewayhighland/DataFiles/LibraryLink- Files/273653.pdf. 50 pp. O’Grady, O.J.T. 2014. Judicial assembly sites in Scotland: Archaeological and place-name evidence of the Scottish court hill. Medieval Archaeology 58:110–142. Ødegaard, M. 2013. State formation, administrative areas, and thing sites in the Borgarthing law province, southeast Norway. Pp. 42–62, In Sanmark, A., F. Iversen, N. Mehler, and S. Semple (Eds.). Debating the Thing in the North I: The Assembly Project. Journal of the North Atlantic special volume 5. Eagle Hill Publications, Steuben, USA. 124 pp. Øystein, E. 2014. St. Olav and the Octagon of Nidaros Cathedral in Trondheim: A Nordic Martyrion? Archaeologiam et Artium Historiam Pertinentia XXVII(N.S. 13):139–160. Owen, O., and S. Driscoll. 2011. Norse influence at Govan on the Firth of Clyde, Scotland. Pp. 335–348, In S. Sigmundsson (Ed.). Viking Settlements and Viking Society: Papers from the Proceedings of the Sixteenth Viking Congress, Reykjavík and Reykholt, 16–23 August 2009. University of Iceland Press, Reykjavik, Iceland. 511 pp. Pococke, R. 1887. Tours in Scotland. Scottish History Society, Edinburgh, UK. 375 pp. Proceedings of the Society of Antiquaries of Scotland (PSAS). 1933. Donations to and purchases for the museum and library with exhibits. Proceedings of the Society of Antiquaries of Scotland 68:1–26. Reynolds, A. 2013. Judicial culture and social complexity: A general model from Anglo-Saxon England. Pp. 699–713, In A. Reynolds and K.P. Smith (Eds.). The Archaeology of Legal Culture. World Archaeology 45.5. Routledge, Oxford, UK. 144 pp. Ritchie, A. 2004. Hogback gravestones at Govan and beyond. The Society of Friends of Govan Old, Glasgow, UK. 11 pp. Robertson, N.M. 1997. The Early Medieval Carved Stones of Fortingall. Pp.133–148, In D. Henry (Ed.) The Worm, the Germ, and the Thorn: Pictish and Related Studies Presented to Isabel Henderson. Pinkfoot Press, Balgavies, UK. 190 pp. Rose, D., Rev. 1791. Parish of Dingwall. Pp.1–20, In J.A. Sinclair (Ed.). Statistical Account of Scotland. 3. London. Available online at http://stat-acc-scot. edina.ac.uk/sas/sas.asp?action=public. Accessed 30 July 2014. Ross, A. 2011. The Kings of Alba c.1000–c.1130. Birlinn, Edinburgh, UK. 245 pp. Royal Commission on the Ancient and Historical Monuments of Scotland (RCAHMS). 1933. Eleventh report with inventory of monuments and constructions in the counties of Fife, Kinross, and Clackmannan. HMSO, Edinburgh, UK. 351 pp. RCAHMS. 1979. The archaeological sites and monuments of Easter Ross, Ross, and Cromarty District, Highland Region. The archaeological sites and monuments of Scotland series 6. HMSO, Edinburgh, UK. 41 pp. RCAHMS. 1982. Argyll: An inventory of the monuments. Volume 4: Iona. HMSO, Edinburgh, UK. 296 pp. Sanmark, A. 2013. Patterns of assembly. Norse thing sites in Shetland. Pp. 96–110, In A. Sanmark, F. Iversen, N. Mehler, and S. Semple (Eds.). Debating the Thing in the North I: The Assembly Project. Journal of the North Atlantic Special Volume 5. Eagle Hill Publications, Steuben, USA. 124 pp. Sanmark, A., and S. Semple. 2008. Places of assembly: New discoveries in Sweden and England. Fornvännen 103:245–259. Semple, S. 2004. Locations of assembly in early Anglo- Saxon England. Pp. 135–154, In A. Pantos and S. Semple (Eds.). Assembly Places and Practices in Medieval Europe. Four Courts Press, Dublin, Ireland. 256 pp. Skene, W.F. (Ed.). 1871. Johannis de Fordun Chronica Gentis Scotorum. Edmonston and Douglas, Edinburgh, UK. 451 pp. Skinner, A.T., and S. Semple. 2016. Assembly mounds in the Danelaw: Place-name and archaeological evidence in the historic landscape. Journal of the North Atlantic Special Volume 8:115–133 Southey, R. 1929. Journal of a Tour in Scotland in 1819. John Murray, London, UK. 276 pp. Storli, I. 2010. Court sites of arctic Norway: Remains of thing sites and representations of political consolidation processes in the northern Germanic world during the first millennium AD. Norwegian Archaeological Review 43(2):128–144. Watson, W.J. 1904. Place-names of Ross and Cromarty. Northern Counties Printing and Publishing Co., Inverness, UK. 302 pp. Whyte, A.C. 2014. Gruline, Mull, and other Inner Hebridean things. The Journal of Scottish Name Studies 8:115–152. Williams, G., and P. Bibire (Eds.). 2004. Sagas, Saints, and Settlements. Brill, Leiden, Netherlands. 153 pp. Williams, H. 2015. Hogbacks: The materiality of solid spaces. Pp. 241–268, In H. Williams, J. Kirton, and M. Gondek (Eds.). Early Medieval Stone Monuments: Materiality, Biography, Landscape. Boydell and Brewer, Woodbridge, UK. 279 pp. Journal of the North Atlantic O. J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 208 Woolf, A. 2000. “The Moray question” and the kingship of Alba in the tenth and eleventh centuries. Scottish Historical Review 79:145–164. Woolf, A. 2006. Dún Nechtáin, Fortriu, and the geography of the Picts. Scottish Historical Review 85:182–201. Woolf, A. 2010. From Pictland to Alba 789–1070. Reprint. Edinburgh University Press, Edinburgh, UK. 384 pp. Endnotes 1Ordnance Survey 1881 1st edition 6-inch map of Ross and Cromarty. 2NAS RH1/2/591/1. 3Ordnance Survey 1880 1st edition 6-inch map of Shetland, sheet LII. 4MS Cotton Claudius D VI, f. 12v (British Library, London, UK). 5NMRS site number NH55NE 4. 6Inverness Journal newspaper 12 February 1813. 7NMRS site no NH55NW 10. The symbol stone is called the Clach an Tiompain and bears the image of an eagle, see NMRS site no NH45NE 6. 8Annals of Ulster 839.9. 9E.g., The Anglo-Saxon Chronicle AD. 869–873, recorded the invasion of the Danish Great Army. See Skinner and Semple 2016 (this volume). 10Annals of Ulster 866.1. 11NMRS site number NH74NW 139, Treasure Trove case TT.159/09. 12Elodil 1476 A.D. 13Scathole 1479 A.D. 14Sweredull 1479 A.D. 15Catoll 1479 A.D. 16Broadpuile 1681, Broadpoole viam ad littus navinum 1686, Broadpool 1696, “Brodpoole the way leading to the Shore” 1707. Walkman in the later 18th century A.D. became known as Wakefield. The -man element of Walkman is also found within the place-name Longman at Inverness and appears to refer to “flat land by a river or estuary”. 17Minute of Dingwall Town Council, 15 November 1833. The place-name Broadpool was last recorded in A.D. 1833 as “the Broad Pool road, from the east end of the Main Street to the sea”. 18NMRS site no NH54SW 7. 19See Owen and Driscoll (2011) on the case for the Doomster Hill as a stepped-mound. 20Annals of Ulster 893.4. 21Annals of Ulster 1085.1. 22E.g. Tynwald, Isle of Man; Tingwall, Shetland Islands; Thingvellir, Iceland. 23Reservato, ad vitam, … montem de Dingwall, juxta villam ejusdem, pro nomine dignitatis ducatus. 24NAS GD305/1/45(e). 25NAS GD305/1/45(e). 26Dictionary of the Scots Language available online at www.dsl.ac.uk, consulted 1 July 2014. 27Dictionary of the Scots Language available online at www.dsl.ac.uk, consulted 1 July 2014, cf. sleikja, ON “to lick”. 28Minute Book of the Presbytery of Dingwall, 18 February 1718. 29NAS B14/1/4/48. 30NAS GD1/436/2. 31“Croftam lie Trinitie-croft cum ejus horto lie Hilyaird ”. 32NAS B14/1/1/6. 33National Records of Scotland ref. no. RHP11591. 34NAS GD93/58, NAS RHP 1474. 35A photograph taken at the time shows the levelling underway and is held in the collections of Dingwall Museum Trust (Fig. 9). 36Further technical details including processing applied to the radar data can be found in the data structure report (O’Grady 2012). 37Further technical information can be accessed in the data structure report (O’Grady 2012). 38Information derived from the British Geological Survey online map viewer. Available online at http://www.bgs. ac.uk/data/mapViewers/msdviewers.html. Accessed on 1 August 2014. 391st edition OS 6-inch:mile 1854 map. 40The name “Doomster” is only recorded as early as the 18th century and is derived from late medieval Scots “judge, court official”. Govan is derived from Brythonic similar to early Welsh meaning ”a small hill”, a language thought to have been spoken in the area before the 10th century A.D. (Clancy 1996: 2–3, Davidson-Kelly 1994: 1–3, O’Grady 2014: 131). 41The mound was levelled in the 19th century. In postmedieval descriptions, it is described as “conical” and ”circular” in shape and may have contained remains of a burial (Davidson-Kelly 1994; NSA 1834–1845: v.6, 690; OSA 1791-99: v.14, 294), evidence played down by previous authors (see Driscoll 1998). This consideration could indicate that the site was a prehistoric barrow, a possibility that would not preclude its later reuse as an assembly mound (see O’Grady 2014: 104, 106, 110, 114). Driscoll’s interpretation of the “ditch” in Paul’s 18th-century illustration of the site perhaps also has not given enough significance to the course of the old burn at Govan that flowed between the church and mound, and was located in this area by excavations in 1996 and 2007 (see Driscoll 1998; Owen and Driscoll 2011: 341, fig.7). See above on alternative location for Tinwald in Drumfriesshire. 42Annals of Ulster 870.6, 871.2. 43Tingwall and Dingieshowe on mainland Orkney, Things Va in Caithness. 44NMRS site no. HY42SW 3, NMRS site no. HY50SW 7. 45Toingal (Lewis), Sunderland (Islay). 46NMRS site no ND06NE 1-2. 47Edin (Cnoc an Rath, Bute), Tinwald motte (Dumfriesshire), and Thingstead (Shetland). Journal of the North Atlantic O.J.T. O’Grady, D. MacDonald, and S. MacDonald 2016 Special Volume 8 209 48Tynwald Hill on the Isle of Man is the best known in the British Isles (Darvill 2004), but see also Thingmote at Dublin (FitzPatrick 2004:28, 45–47) and local thing mounds in Sweden and Norway (Brink 2003, 2004; Sanmark and Semple 2008).