When were weapons banned at the Thing assemblies in Norway?

What exactly happened at early things is sparsely recorded, if not to say for really earlier times wholly non-existent as traditional sources. Without written sources, things get complicated.
But we have hints. Among those is that for a medieval perspective this civilising tendency was apparently comparably late, and parallel in temporal development with the increasing power of a centralised monarchical organisation.

Increasing power to the king, decreasing power to free and armed man. Putting an exact date to all things is probably not the correct question, as this was not only gradually done – from restrictions on use and control, over to ban – but also at different speeds in different places.

A first glimpse on the proceedings in question:

From what we know, it seems obvious that in early Scandinavia it was the custom to make a noise with weapons at thing assemblies for expressing opinions, thus the divisions of wapentakes in the Danelaw inform us, but the custom is also mentioned in the much later Magnus Lagabøter’s Law (landslo ̨g) (I:5; NGL 1: 409), where it says that a verdict is not legally valid unless the people on the thing assembly, who stand outside the marked-out and hollowed-out area where the judges sit, lo ̨grétta, give their consent to the verdict by rattling or raising their weapons in the air (vápnatak or þingtak).
_{— Stefan Brink & Neil Price (eds): "The Viking World", Routledge: New York, London, 2008.}

Note that the Magnus Lagabøtes landslov is commonly dated to 1274–1276.

How this was made to look less warrior like is a process played out over quite a long time and in different places on different levels:

Even though one should be careful not to infer ex silentio, the fact that Varlo is not recorded as a thing site in medieval documents might indicate that the assembly was moved from Varlo to Berg. The transferral of the assembly may have happened before the mid-14th century, which is when the first diploma indicating legal functions at Berg is dated (DN III:299; 300, 1358–9; Lange and Unger 1853:243, 244).

After 1273, documents started being more commonly produced at thing meetings, especially concerning cases with physical injuries (RN II:112; Bjørgo and Bagge 1978:70).
[…] As stated above, thing meetings at Berg are repeatedly referred to as weapon things, designating assemblies on the skipreiða level. […] The Law of the Gulathing states that only free men of full age can attend the assembly, and that the handling of weapons should be controlled (G 309; Eithun et al. 1994:166–167). It is likely that the right to carry weapons was regulated also in earlier times (Hofset 1981, Martens 2003). […] The weapon thing at Berg perhaps suggests a family of an extraordinary position and higher status than others. At least, being one of the king’s “faithful men” may suggest that this farmer had greater influence over decisions at the assemblies than others (Moseng 1994:95), while perhaps also giving the king some control over the thing organization at the lower level in Eiker.

Maybe changing the location of the thing site, and perhaps also the function, is a consequence of the king gaining ownership of the farm of Sem? The date coincides chronologically with the period when the control of weapons at the assemblies had, or was in the process of, being transformed into a tax (Andersen 1982a:346–359). Consequently, it is likely that taxes in kind were collected at the assembly of Berg and moved to Sem for storage and use.

Then again, why not move the assembly to the royal farm itself? The relation between assembly sites, chieftain’s and royal farms is not properlyly understood, and several different situations have been observed (e.g., Andersen 1977:225, Reynolds 1999:65–110, Sanmark 2011, Turner 2000).

In northern and southwestern Norway, there seems to be a close connection between chieftain’s farms and courtyard sites, interpreted as assemblies on neutral ground where the landowning elite could meet for political and religious activities (Storli 2000, 2010).

This desire for a neutral site could perhaps explain the location of the weapon thing at Berg, not directly located on the royal farm itself, as the “farmer” might have functioned as a royal official with close connections to the king and his administration, while at the same time also serving as local representative and sanctioning the law for the local population. The royal of cials were dependent on the thing participants’ cooperation, whilst looking after the king’s interests in the local communities, for example the collection of taxes. The thing thus became an arena for cooperation between the royal representatives and the farmers (Dørum 2004:387, Imsen 1990:193–198).

Previous researchers have presumed that the king took strong control over the assembly organization after the mid-12th century, especially through his representatives. The development of the administrative structures and royal administration discussed in the regional analysis seems to support this view. Most likely, the king gained more control over the thing organization by drawing the activities closer to new urban centers and central administration, at least at the highest administrative levels. This paper has examined how power may have been extended down to the lower levels.

In Eiker, three of the suggested thing sites had the same location across the centuries, which seems to imply a continuity of the political system. However, as shown, the thing system was not static, and changes both in the thing site location and administrative districts could happen. The main skipreiða thing in Eiker was moved to Berg in the 13th century, probably to be closer to the royal farm of Sem.

This move could probably be related to increased royal control over the local thing institution and the judicial life in the districts by the beginning of the 13th century. Nevertheless, the thing was an arena for the general public in the local communities, and especially before the 15th century, this was a result of cooperation between the royal representatives and the farmers. The thing sites gradually became an instrument for communication and cooperation between the local communities and royal authority.

This is, however, a study of only one skipreiða, and not necessarily representative for the Borgarthing law province as a whole.

_{— Marie Ødegaard: "State Formation, Administrative Areas, and Thing Sites in the Borgarthing Law Province, Southeast Norway", Debating the Thing in the North I: The Assembly Project, Journal of the North Atlantic Special Volume 5:42–63, 2013. (PDF)}

Note the vague description of "weapons control". This is all in the range of 'do not bring them', 'lay them down beforehand', 'take them to the assembly but do not draw them', 'bring them along but during session keep them secured with a friðbönd (peace straps)'.

The mentioned sources looking closer at weapons control at things

• Eithun, B., M. Rindal, and T. Ulset (Eds. and Trans.) 1994. Den eldre Gulatingsloven. Norrøne tekster 6, Riksar- kivet, Oslo, Norway. 208 pp.
• Hofseth, E.H. 1981. Loven om våpenting sett i lys av arkeologisk materiale. Pp. 103–118, Universitetets oldsaksamlings årbok 1980–81. Universitetets old- sakssamling, Oslo, Norway. 211 pp.
• Martens, I. 2003. Tusenvis av sverd. Hvorfor har Norge mange ere vikingtidsvåpen enn noe annet europeisk land? Collegium Medievale 16:51–66.

That weapons were present during a 'weapons thing' – originally – should be self-evident. The Gulathingslov prescribes this as follows:

Vm vapna þing. 309 Hvervetna þess er vapna þing scal væra. þa scal armaðr æða lendr maðr segia til of haust oc hava þing of vár. scolo aller menn þat þing sökia frialser oc fulltiða. ellar ero þeir vittir .iij. aurum hverr þeirra. Nu scolo menn vapn sin syna sem mælt er i logum. scal maðr hava breiðöxe. æða sverð. oc spiot. oc skiolld þann at versta koste er liggia scolo iarnspengr þriar um þveran. oc mundriði seymdr með iarnsaumi. Nu liggia aurar .iij. við folkvapn hvert. Nu scolo böndr fa til a þofto hveria tvennar tylftir orva oc boga einn. giallde eyri firi odd hvern er missir. en .iij. aura firi boga.

_{— Jvfr. Cap. 307. F. VII. 3. Cap. 308. F. VII. 5. Cap. 309. F. VII. 13. 15. Hvárt skip synizt fört. æða eigi. — From Norges gamle Love/Den ældre Gulathings-Lov, this text published 1847, but dating the original written source is contested, from 11th–12th century.}

Partial translation:

Whenever a weapons thing is to be held, […] All free and adult men must attend it, or each of them is punishable by three Øre. There the men shall show their arms, as the law requires. The man shall have a broad axe or sword, a spear and a shield, over which at least three iron bands are laid crosswise and the handle is nailed on with iron nails. A penalty of three øre is imposed on each weapon [if it is missing]. In addition, the bindings shall provide two dozen arrows and a bow for each oar. For each missing arrow they shall pay one Øre and three for the bow.
_{– own translation, perhaps too freely or incorrect, if anyone speaks that language or has at the ready a better translation to English, please improve.}

And again Magnus Lagabøtes, as his correspondingly bylov (city law), in this case for Bergen, where the thing already wasn't held under open air like in more rural areas but in a hall, differs a little a litte later:

The inhabitants of the town, like all Norwegians, were liable to military service and had to ensure that they were properly armed according to their wealth. The unemployed wage earners had to buy an axe in the 1st year, a shield in the 2nd year and a spear in the 3rd year. Once a year during Lent, an appeal to arms (weapons thing) was carried out, where the mayor and the councillors, together with men from the royal retinue who were familiar with weapons, convinced themselves of the proper condition of the weapons. However, Cardinal Nicholas Breakspear had already demanded a ban on the carrying of arms in the trading town when he elevated the diocese of Nidaros to the status of archbishopric in 1152. This ban was repeated again and again, which suggests that it was not very successful. Only Sýslumaður and Gjaldkeri were allowed to carry arms.
— src

If despite that a single number is sought for precision, with caveats we may offer this on "weapons banned at Norwegian thing assemblies":

The emergence of the weapon-thing may have been closely connected to a development where kings established a monopoly of violence in a given region. All free male adults were obliged to attend the weapon-things, which were not representational things such as the quarter-thing. The men were obliged to participate in the defence system, if called upon, and the purpose was to regularly ensure that their weapons were in accord with the stipulations of the law. According to the Gulathing law, royal officials were required to announce in the fall the weapon-things to be held in the spring (G 309).
In 1189, a prohibition against bearing arms at things and in churches emerged (NgL I, 409; RN I 198).
The weapon thing was transformed to a taxation thing. Nevertheless, the kings’ acquisition of the military judicial functions may have been an important motive for the abandonment of courtyard sites. Let us examine this prospect more closely.
_{— Frode Iversen: "Houses of Commons, Houses of Lords? The Thing on the Threshold of Statehood in Rogaland, Western Norway in the Merovingian and Viking Ages", in: Irene Baug, Janicke Larsen and Sigrid Samset Myglan (eds): "Nordic Middle Ages – Artefacts, Landscapes and Society. Essays in Honour of Ingvild Øye on her 70th Birthday", UBAS University of Bergen Archaeological Series 8, Bergen, 2015. p175–192. (hdl)}

Why do these dates seem to contradict each other fundamentally?
One other aspect to observe is that even in Magnus' time the weapons are at the same forbidden and needed?

One solution to this problem is not only that there were simply different kinds of things.

Another solution is that the strictest prohibitions at first did not apply to the thing-place as a whole, but only to the inner circle, a holy sanctum, the vébǫnd where absolutely no weapons were allowed (also compare first quote here). So the inner structural and spatial organisation of individual thing places seems to have played another role.

The sacred bonds (vébǫnd) thus separated the sacred space from its profane surroundings at the thing-place in both Frosta and Gula, and probably at many other “sacred” assembly places in Scandinavia. All those who entered such areas had to leave their weapons behind, and the people standing inside the bonds were treated as inviolable. […] They acted in the opposite way to a just ruler, who must protect the court-circle, since it was regarded as a sacred space. The rule and order that prevailed at such places must have been in the interests of the ruling power. The sacred space of both sanctuaries and thing-places constituted the most important arenas where the rulers could manifest their own power, by showing their authority, wealth and glory in public and in ceremonial circumstances. The ritual restrictions and taboos were therefore actually in favour of the ruler at such gatherings. When he appeared on the public scene he did not need to feel physically threatened by enemies. No weapons were allowed to be carried at such places. In Egils saga 25, for instance, we read that “it is a custom here to meet the king unarmed”. In this passage King Haraldr was attending a ceremonial feast (var á veizlu) at Voss in Hǫrðaland, which took place in a hall (stofa), probably a ritually protected place.

In the warlike society of Viking Age Norway it must have felt safe for the leaders to inhabit places where ritual and religious restrictions prevailed, i.e. the sacred places. If someone attacked the ruler with weapons at such a place, he or she not only offended this person and his office, but also the religious and judicial rules which prevailed at such space. These regulations were estab- lished and sanctioned by old custom (siðr). Indeed, any such violation had to be considered a capital crime, since it was directed against the society and perhaps even the cosmos itself. If the ruler was attacked when attending the sanctified grounds, for instance in the ceremonial building or at the sacred meadow, the anger of the mythical powers was also aroused. Such a violation could lead to disturbances in the cosmos, because the sanctuary was regarded as the mythic landscape on earth. Therefore the taboos and ritual restrictions at cultic sites in Trøndelag could be seen as a strategy connected to the rulers and political power, not only in this region, but also elsewhere in Norway, Iceland and Svetjud.
_{— Olof Sundqvist: "Sacred Sites, Ritual Restrictions and Religious Strategies for Rulership", in: Numen Book Series, Vol 150, "An Arena for Higher Powers", Brill: Leiden, 2016. DOI}

A very nice analysis for this connection of the kings as the actual law breakers by disenfranchising their subjects from their old customs – with a focus on weapons in or around holy places – is found in

— Olof Sundqvist: "Vapen, våld och vi-platser. Skändande av helgedomar som maktstrategi i det vikingatida Skandinavien", in: Håkan Rydving & Stefan Olsson (eds.): "Krig och fred i vendel- och vikingatida traditioner", Stockholm University Press: Stockholm, 2016.
_{(PDF easily found online, but legality unknown, so no link).}