Specific Phase names for the Germans

[Thalatta]

I wasn't satisfied having just said "regarding something in particular": I mentioned Kroonen before because he gives þurpa meaning as "crowd", "which anyway blatantly contradicts your “singular first” interpretation". Then I ended up finding quite some interesting stuff:

In the 2023 dissertation "Unde vēnistī? The Prehistory of Italic through its Loanword Lexicon", by A. M. Wigman, one of Kroonen's PhD students, one can find:

Quote

Sabellic attests to a verbal root *treb- ‘to dwell’, with a root noun *trēb- ‘house’ with cognates in PCelt. *trebā- ‘settlement’ (e-grade), PGm. *þurpa- ‘crowd, settlement’ (zero-grade), and PBalt. *trōb- (o-grade lengthened by Winter’s Law)

It doesn't seem Kroonen disagreed much with his own student for that to end in the PhD thesis… not that we didn't know what "crowd" seemed to support already.

Later on, it's stated that:

Quote

It is interesting to note that most of the Germanic comparanda, where they contain a diagnostic phoneme, can be shown to have entered Germanic before the operation of Grimm’s and Verner’s Laws. It is a pre-Proto-Germanic stage when their consonantism can be reconstructed to match that of the other comparanda most closely

And the relevant row compares Proto-Germanic *þurpa- with Proto-Celtic *treb- (a footnote states that "a note of caution is that it cannot be ruled out that reconstructions involving the operation of Verner’s Law might be misinterpretations", showing that, although a possibility, this is not a mainstream view, unlike what's in the main text).

Finally, it's stated that many things there come "as no surprise if Grimm's Law operated in the mid-first millennium BCE", which shows this is already considered, and bears no major problem towards the reconstructions provided, making evident that this is not an issue:

On 05/11/2026 at 2:07 AM, Genava55 said:

deriving *þurpą from *treb- requires accounting for an unexpected voiceless stop (*p) where a voiced aspirate or plain voiced stop (*b / *bʰ) would regularily be expected under Grimm's Law

Here, why are you only considering one change? There are more as part of Grimm's Law, as From Proto-Indo-European to Proto-Germanic, by D. Ringe, states: "by the second change subsumed under Grimm's Law PIE voiced stops became PGmc voiceless stops". This turns *b into *p, as expected, and the third example in that very same page is in fact:

Quote

PIE *treb- ~ *tr̥ b- ‘building’ (cf. OIr. atreba ‘(s)he dwells’; secondary zero grade in Lat. trabs ‘beam’) in PGmc *þurpą ‘farmstead, village’ (cf. ON þorp; Goth. þaúrp ‘field’);

Which also includes village as one of the meanings of *þurpą.

[Genava55]

On 06/11/2026 at 12:11 AM, Thalatta said:

Honestly, you need to find some integrity at some point and just be able to say "you know what, I was wrong with that", instead of throwing false accusations and keeping muddying the waters because your baseless arrogance doesn't allow you to recognise when it's resoundingly obvious that you've been completely cornered.

I have no reason to acknowledge any error in this case. It is an established fact that the Épinal-Erfurt glossary mentions the word 'tuun' in only two contexts, and that its only direct translation in this glossary is the Latin 'cors'. There is no explicit link in this glossary between the word 'tuun' and the concept of a village. There is, therefore, no reason to assume that 'conpetum' denoted a village solely based on the glossary. The dictionary I am criticizing (Murray’s New English Dictionary) uses the Épinal-Erfurt glossary to claim that both 'thorp' and 'tuun' meant 'village,' with the dictionary using one term to explain the other. I demonstrated this by examining both entries and tracing them back to their earliest sources, which refer to one another, a clear case of circular reasoning. They could have employed other arguments, but they did not. Consequently, my critique of this academic source is legitimate.

If anyone wishes to argue that 'thorp' or 'conpetum' meant something along the lines of 'village' during that period, they must absolutely support that claim with an argument based on a body of evidence. The same criticism applies to the online edition of the Épinal-Erfurt Glossary that you're using. It offers a translation, but there's no explanation or justification. It's just a raw proposal. Once again, based solely on the glossary, it is difficult to argue that 'conpetum' and 'thorp' referred to a village. I am simply bringing this issue to light and offering an alternative interpretation that is entirely consistent with the primary sources (see below). 

On 06/11/2026 at 12:11 AM, Thalatta said:

Yet what you did is just to shove those sources when you didn't even properly read them. Calling showing you their actual contents "appeal to authority" or "mindlessly parroting" is not going to fool anyone, so you might want to change this old defensive trope, it got a bit repetitive and boring already. Just start by reading your own sources properly, in light of what the point of the discussion actually is, and maybe that will be a bit more productive.

I think you are misunderstanding what I meant by an appeal to authority. My objection is not that you cite academic sources. I have cited academic literature myself throughout this discussion, and I fully recognize its importance. The problem is that you repeatedly cite their conclusions as if the authority of the source were enough to end the discussion, without explaining how those conclusions were reached or addressing the criticism being made against them.

Once again, I want to emphasize this point. What is starting to frustrate me is not the fact that you cite academic sources. What I object to is the way you repeatedly present the conclusion of a dictionary or a scholarly work as if its authority were enough to settle the question, without examining the evidence or explaining the reasoning behind that conclusion.

This is what I meant by an appeal to authority. I was not suggesting that citing specialists is inherently illegitimate. The problem arises when the citation is used as a substitute for an argument. Saying that a dictionary reconstructs 'þurpą' as 'village' establishes that this is an accepted interpretation, but it does not demonstrate why that interpretation is correct, nor does it answer a criticism of the evidence on which it may be based.

That distinction is particularly important here because the interpretation of the sources is precisely what we are debating. If the reconstructed meaning 'village' was inferred partly from later West and North Germanic evidence, it cannot simply be projected back onto earlier or ambiguous attestations and then used to validate the original reconstruction. That creates a real risk of circular reasoning. The conclusion is treated as a premise, and the evidence is then interpreted through the conclusion it is supposed to establish.

The same applies when an editor translates an ambiguous gloss as 'village' or 'hamlet'. That translation is relevant and deserves to be considered, but it remains an interpretation of the primary evidence. It does not make alternative readings illegitimate, especially when the words involved have broad semantic ranges and when other early attestations point toward agricultural land, estates, farms or dependent settlements.

What I find unhelpful is the repeated attempt to short-circuit this discussion by answering a criticism of the sources with the authority of the very sources being criticized. Academic works are not beyond examination. Their conclusions can be questioned by looking at their premises, their choice of evidence and the semantic assumptions behind their translations. That is not a rejection of scholarship. It is how scholarship is supposed to be discussed.

If you want to defend the interpretation 'village', then explain why the primary evidence supports it better than the alternatives. Show why the Gothic evidence should be considered innovative, why the agricultural and estate-related meanings should be secondary, and why the later collective meanings represent inheritance rather than parallel development or semantic expansion. Simply repeating that a dictionary or an editor adopts that interpretation does not answer those questions.

I am perfectly willing to engage with academic sources, including sources that contradict my position. What I do not accept is the suggestion that questioning their reasoning is somehow illegitimate, or that citing their conclusion is enough to close the debate

On 06/11/2026 at 12:11 AM, Thalatta said:

It's you who keeps "dodging the arguments that contradict you", even when using references first put forward by you. The real question here is, why would anyone give more weight to entries from decades old general purpose dictionaries instead of very recent specific studies on the actual matter?

The source I am referring to is a primary source. The Harley Glossary is an 11th-century Anglo-Saxon manuscript (British Library MS Harley 3376) that pairs Latin headwords with Latin and Old English explanations.

‘competum .i. uilla, uel þingstow, uel þrop’ (from The Harley Latin-Old English Glossary published by Robert T. Oliphant)

Here 'competum' is translated by 'villa' (farm, estate), 'þingstōw' (place of assembly) and by 'þrop' (or 'thorp')...

I'm just pointing this out to show you that even several centuries later, 'conpetum' wasn't translated as 'town' or 'village'. It is translated as a place of assembly, a farm or an estate, and 'thorp'. This supports my point regarding the Épinal-Erfurt glossary: the equivalence between 'thorp' and 'tuun' might instead stem from the concept of an enclosure or courtyard.

Which once again is supported by the explanation made by Isidore of Seville:

Original: Conpita sunt ubi usus est conventus fieri rusticorum; et dicta conpita quod loca multa in agris eodem conpetant; et quo convenitur a rusticis.

Translation: Conpita are the places where farmers usually gather; they are called conpita because many country roads converge there (conpetant) and because that is where the farmers meet.

So here I am presenting an argument based on several sources simply to show that there may be another very credible and legitimate interpretation. 

And the idea of challenging the conventional dictionary interpretation, that 'thorp' did not originally refer to a village but rather to a farm or an agricultural estate, is a view supported by other researchers. 

For example, while you cited Ekwall for 'tuun', he himself wrote the following regarding 'thorp':

OE þorp, þrop is a rare word, and its meaning is doubtful. It was certainly used in the sense ‘farm’, possibly in the sense ‘hamlet’. There is no reason to suppose that it meant ‘village’. [...] The original meaning of þorp was ‘newly reclaimed land, new settlement’. It should not be rendered by ‘village’, but rather by ‘farm’.

[Thalatta]

Thanks for answering like a reasonable person. I agree with what you say regarding sources. I just think you are challenging the current sources by using older ones that are actually being criticised by the former in the first place (Oliphant and Ekwall published over 60 years ago, let alone other sources you have quoted). You have to recognise that as circular reasoning. I'll try to demonstrate this as explicitly as possible, while trying to answer what you ask.

You wanted to shut down the "hamlet or village" meaning and imply circular reasoning from the current EE experts with your statement characterising the "village" entry as "completely illogical". You also said "tuun at this time still meant enclosure", and (using the "Compitum = Tūn = Throp" equivalence), "throp in the Épinal-Erfurt is obviously not of village". This line of reasoning is unsubstantiated, since the "village" meaning appeared early during Saxon colonisation, before the EE. You hopefully now agree with that, given that you said "this reflects both the semantic shift toward the meanings of "homestead" and "village", which, according to Ekwall, emerged early on without distinguishing between the two".

Still, this only proves that the 2016 (reviewed in 2022/2023) "village or hamlet" translation is possible, not that it's correct yet. How to get closer to that? Let's look at the current evidence, particularly if it analyses the older evidence. I'm going to use another of your sources: Thorps in a Changing Landscape, from Richard Jones, David N. Parsons and Paul Cullen, from 2011, containing the extremely relevant "Appendix 1: Throp in Anglo-Saxon glosses", which you used to heavily recline on compitum to justify thorp just being a "gathering place". You quoted:

On 10/05/2026 at 11:27 PM, Genava55 said:

throp appears in association with Lat compitum, which primarily means ‘cross-roads’. In one instance it is additionally combined with OE thing-stow ‘meeting-place’. These senses are rather different from what is generally suggested by Germanic cognates and later English evidence, but they have been admitted into the dictionary definitions of throp, have been sought in some place-names and have coloured discussions of the Germanic term’s ultimate etymology

and then you said:

On 10/05/2026 at 11:27 PM, Genava55 said:

The current theory regarding the Anglo-Saxon period is that “þrop” referred to an estate acting also as a gathering place for peasants who collectively managed land, and that there was a semantic shift in its meaning to refer first to the entire community and then to a village.

Which I don't find convincing given the current evidence. So I went to your source, and found that then it says:

Quote

closer inspection of the glosses, however, suggests that it is not at all certain that either ‘cross-roads’ or ‘meeting-place’ was ever a straightforward translation of OE throp

and:

Quote

it is not enough to say that throp in this complex is given as an equivalent to Lat compitum ‘cross-roads’. Rather, throp and tūn were together originally associated with compitum. Since tūn is an enclosure and settlement term – the semantics of which are fairly well understood and do not encompass ‘cross-roads’ – it is necessary to question how the original glossator understood the Latin word: this will inevitably have implications for our understanding of throp

and then mentions the "plausible guess" that all could have come from "qui alieni a ciuitate Dei ex locorum agrestium conpitis et pagis pagani uocantur", that is, "those who, alienated from the kingdom of God, are called ‘pagans’, from the conpitis and pagis of rural places", concluding that:

Quote

certainly, if this were the source of the original gloss it is easy to see how terms for rural settlements might have been provided for the Latin words which actually mean ‘cross-roads’ and, perhaps, ‘districts’ respectively. Indeed, in this context throp and tūn, straightforwardly denoting small, agrarian settlements, look rather more appropriate than a term meaning ‘cross-roads’

And you saw but omitted all that, because that's followed by the sentences you quoted from Isidore of Seville regarding what a compitum is, and then even said:

On 10/05/2026 at 11:27 PM, Genava55 said:

So, once again, it’s a place, a courtyard, or an enclosure. It makes so much more sense with the word "tuun" that I don't understand how you can't see it. If, in the same glossary, the word "tuun" is also translated as "enclosure," it cannot mean anything else in this context.

when actually that source, just after the Isidore of Seville statements, adds:

Quote

Isidore correctly relates the word to the verb competere ‘to meet, come together’. Although the primary sense of the noun would appear to be simply ‘junction, cross-roads’, subsequent development of these junctions as meeting-places for the rural community is readily comprehensible, and this usage was presumably in Orosius’s mind when he wrote his passage. On various counts, then, it is unlikely that the Anglo-Saxon who glossed compitum with tūn or throp was searching for words that meant ‘cross-roads’. Had that been the intention then tūn, at least, would surely not have been chosen. And if the source of the Latin word was Orosius’s text, or something similar, then ‘cross-roads’ is not really the operative sense. In a context where the author is describing the rural haunts of unsophisticated people, compitum may easily have been misunderstood, or partially understood, as referring to small rural settlements

And the source continues a bit regarding these things, stating that glosses

Quote

tend to suggest – especially in the explicit pairing of throp with tūn – that the intended sense here, as in place-names and later records of English, is basically ‘(small, rural) settlement’.

And that chapter ends in that page, not changing the conclusion that both throp with tūn refer to "small agrarian settlements", and compitum is not a "straightforward translation", but probably "misunderstood, or partially understood, as referring to small rural settlements" by "unsophisticated people" (to which I independently hinted before with another source you didn't find convincing, even after you had read that).

This is why I say you can't just give more weight to older entries lacking in analysis than to current relevant discussions directly criticising them. And please notice that your quoting highly misrepresents what's being said: you end up concluding the opposite of what's actually said. I could go on ranting about this, but I'll refrain because you were respectful in your previous answer. They clearly explained what you quoted in order to promptly criticise it. I leave "hidden" screenshots of this nice 6-page long appendix for anyone to read it:

Spoiler

 

There are many other things worth checking from that source about thorps which directly imply them being more than a single farmstead, for example this quote explicitly stating that that would just be the case for quite derelict ones:

Quote

Danelaw thorps first recorded before 1500 (264 out of 576) are now completely lost or are represented by little more than a single farmstead

One has to bear in mind that certain thorp characteristics don't necessarily apply to þurpą, being a cognate many centuries after the fact. All this focuses only on supporting the EE translation being a small settlement.

 

I think it's then clear that there's no problem with the current evidence taking throp and tūn as a settlement larger than a single farmstead. This association can be tracked all the way back to þurpą, more recently shown in Kroonen's PhD student use of "settlement" (its association with "crowd" can also be found in other languages, that source mentions it for OHG, and it happens in Spanish with "pueblo", of course independently), and clearly present in the Celtic branch through PIE treb, a canonical example of Grimm's Law, as I've shown.
 

In summary (and if you think I'm misrepresenting something, briefly point it out), because again, all has to be seen at the same time to make the actual point:

Þurpa: at least the meanings crowd, settlement and hamlet are justified (by your own sources). While (full-blown) village might be an exaggeration, it is irrelevant for Phase I, as is the additional farmstead meaning, considering the nature of these settlements, and how Romans saw them. Gothic cannot be reliably used for this word in particular since there's only one attestation. You tried to argue that the settlement meaning only came much later in Old High German and Old Norse, but earlier Old English attestations prove that this cannot be the case, fact reinforced by the many languages presented in the Celtic branch, both reconstructed and attested, from where a size estimation of this larger than a farmstead settlement was also obtained. Even when for a few Celtic languages þurpą cognates are less settlement-like, they are still more so than their corresponding haimaz cognates, an extremely important point. All this shows that þurpą early on meaning only a single farmstead seems unlikely. Its meaning refers more often to a small settlement.

Haimaz: at least the meanings home, homely and homestead are justified (by your own sources). Gothic has a stronger case here, although it translates haimaz cognates as country at least once, inverts wīhsą cognates urbanistic order, and all this is centuries after the fact. You tried to argue for it meaning settlement using Old High German, but this is ten centuries after the fact. Haimaz cognates ended up meaning settlement as in "someone's home", but this was not its original meaning, not only because the homely meaning is ever present and ambiguating in the Germanic branch, but because it's astoundingly dominant in the Celtic branch, showing that by the time PIE split into these two branches there was probably no settlement meaning yet, and the lack of any horizontal transfer hints to a rather late than early development, but it cannot be said at which point between Proto-Germanic and Gothic this meaning appeared. All this shows that haimaz early on meaning mainly settlement seems unlikely. Its meaning refers more often to a home.

In conclusion, I do think þurpą is a better candidate than haimaz for a Phase I Proto-Germanic settlement, and that the accepted reconstructions are properly backed up by the latest evidence, which contains plenty of thoughtful criticism of the older evidence, as shown.

Edited November 7 by Thalatta

[Genava55]

19 hours ago, Thalatta said:

certainly, if this were the source of the original gloss it is easy to see how terms for rural settlements might have been provided for the Latin words which actually mean ‘cross-roads’ and, perhaps, ‘districts’ respectively. Indeed, in this context throp and tūn, straightforwardly denoting small, agrarian settlements, look rather more appropriate than a term meaning ‘cross-roads’

19 hours ago, Thalatta said:

tend to suggest – especially in the explicit pairing of throp with tūn – that the intended sense here, as in place-names and later records of English, is basically ‘(small, rural) settlement’.

19 hours ago, Thalatta said:

One has to bear in mind that certain thorp characteristics don't necessarily apply to þurpą, being a cognate many centuries after the fact. All this focuses only on supporting the EE translation being a small settlement.

 

I think it's then clear that there's no problem with the current evidence taking throp and tūn as a settlement larger than a single farmstead.

I have never tried to conceal the fact that the authors use the word settlement. On the contrary, I provided the reference openly. Furthermore, in case you hadn't noticed, Ekwall also uses the term 'settlement' while associating it with the word 'farm'. The disagreement is not about the presence of that word, but about what this very broad term means in this particular context and what conclusions can legitimately be drawn from it.

The Cambridge Dictionary, for example, defines settlement simply as "a place where people come to live or the process of settling in such a place". Nothing in that definition necessarily requires several independent households, a cluster of houses, or anything resembling a village. A large isolated farm or estate can therefore constitute a settlement in its own right, even if it contains only one principal residential building. It may also include barns, stables and other agricultural structures, but those additional buildings are not what makes it a settlement. The essential point is that it is a distinct place where people have established themselves and live.

This distinction matters because you repeatedly treat settlement as though it necessarily implied a collective residential community larger than a farmstead. It does not. A farmstead, an agricultural estate, a dependent holding, a hamlet and a village can all be described as settlements. The word itself does not tell us how many households were present, whether the buildings were nucleated or dispersed, or whether the site constituted an independent community. I fully understand that the word can be confusing, since I myself have made this mistake in the past, especially because English is not my native language. However, in research, this word refers to a concept distinct from the concept of the number of buildings. As P. J. Duffy notes in the "Historical Geographies, Rural" entry of the International Encyclopedia of Geography, "Historic rural settlement patterns occur in many different forms, from single dispersed farmsteads, through a range of agglomerations of simple farm clusters and hamlets to much more complex nucleations around greens, churches, or castles."

If I may give two examples of academic researches along these lines: https://uni-tuebingen.de/en/faculties/faculty-of-humanities/departments/department-of-history/institutes/ancient-history/research/lycia-project/research-areas/tower-and-isolated-farmsteads/ & https://architectus.pwr.edu.pl/files/numery/75_05.pdf

In the book (Thorps in a Changing Landscape), I understand settlement primarily in the sense of a newly established farmstead, agricultural estate or dependent rural holding, rather than as an unambiguous synonym for 'village'. This reading is not arbitrary. From the beginning, the authors emphasize the agricultural context of the word and repeatedly connect thorps with farms, estates, cultivated land, soil types and changes in agricultural organization. Their study is not simply an attempt to prove that thorp meant 'village'. It investigates what distinguished these particular rural establishments from places bearing other habitative name elements.

The authors explicitly challenge the standard explanation that thorps were merely small villages attached to more important places. Their argument is that this conventional label does not adequately explain what made these places distinctive. They instead examine their subsidiary status, their relationship with older centres, their position in the landscape and their role in the reorganization of arable farming. In other words, their principal concern is not whether a thorp had crossed some demographic threshold separating a farm from a hamlet or a village. Their concern is the specific economic, territorial and agricultural function that justified the use of the name.

Their archaeological evidence admittedly shows that some thorps eventually developed into nucleated settlements with several holdings, regular plots or dependent tenants. I have never denied that. But this does not establish that 'village' was the original lexical meaning of the word. It shows what some places bearing that name became during a period of major agricultural and social reorganization.

That chronological distinction is essential. A newly founded farmstead or dependent agricultural establishment can attract additional households and gradually become a hamlet or village while retaining its original name. Place names regularly survive major transformations in the size, structure and function of the places they designate. The fact that a medieval or modern thorp was later a village does not prove that the name originally denoted a village when it was coined.

Indeed, the authors note that thorp and throp names can point today to villages, hamlets or individual farms. That diversity is exactly what we would expect if the term had been applied to a type of rural establishment whose later development varied from one site to another. Some remained farms, some disappeared, and some developed into larger and more nucleated communities. The later outcomes cannot simply be projected backwards and turned into the original definition.

It is also important that the authors situate the creation of these names within changes in agricultural organization and the process of village nucleation. That wording implies a historical development, not an original equivalence between thorp and 'village'. If thorps participated in or contributed to a process of nucleation, it does not follow that they were already fully developed nucleated villages at the beginning of that process.

Nor do the authors simply reproduce the conventional model of thorps as dependent hamlets. They explicitly ask us to move beyond that characterization and investigate the particular features that made them distinctive. The book even states that, rather than dismissing them as 'secondary settlements' or dependent hamlets, we should examine their specific characteristics and their role in transforming farming practices. This is much closer to a functional and agricultural interpretation than to your claim that the authors straightforwardly support 'village' as the original meaning.

I am therefore not arguing that settlement can only mean 'farmstead', nor am I claiming that every historical thorp consisted of one household. My argument is that settlement is far too broad to substantiate your interpretation. If thorp is glossed as settlement, that remains compatible with a farmstead, an agricultural estate, a dependent rural establishment, a hamlet or a village. The word 'settlement' alone cannot tell us which of these was the original referent.

The authors' overall position appears considerably more nuanced than the one you attribute to them. They treat thorps as relatively small rural establishments, often new or subsidiary, closely connected with cultivated land and the reorganization of arable agriculture. Some developed into compact communities with several holdings, while others remained individual farms or disappeared. Their later, broader use as a designation for small rural settlements should not automatically be projected back onto the earlier stages.

In short, both the book and the article do not establish that thorp originally meant 'village'. They suggest that thorps were distinctive rural establishments whose agricultural role, dependent status and development contributed to the long process by which parts of the English countryside became more nucleated. The village is therefore better understood as one possible outcome of that development, not as the only meaning already contained in the word from the beginning.

The use of settlement consequently does not contradict my interpretation. It encompasses it. If you want to demonstrate that the term originally referred specifically to a collective settlement composed of several independent households, rather than to a farmstead, estate or dependent agricultural establishment that could later develop into such a community, you need evidence beyond the English word settlement. The source you cite does not, by itself, establish that conclusion.

This also offers a different explanation for the earlier association of thorp with tuun. The two words did not need to be exact synonyms. Glossaries frequently group words because their referents overlap, not because their meanings are identical. A tuun could be an enclosed holding, and a thorp could be a rural establishment occupying such a holding. Their equivalence in one gloss may therefore reflect a shared physical or functional feature rather than a shared meaning of 'village'. I nevertheless think that the equivalences found in these glossaries follow a coherent semantic logic, as the authors themselves suggest. In my view, the association between competum, tuun, thorp and þingstow may rest on the shared notion of a physically delimited rural area, used for cultivation, habitation, landholding or local assembly. These words need not have been exact synonyms. They may simply have described different aspects or functions of the same kind of agricultural site.

I hope this lengthy explanation helps you better understand my position and my perception of the literature on the topic. Personally, this kind of discussion is precisely the heart of my argument. The study of the meaning of words goes beyond simple etymological reconstruction. In my view, the book and the article effectively emphasize the importance of reframing the evolution of the word 'thorp' within the historical context of the Middle Ages and the many changes that transformed the agricultural world. As Stefan Brink says in the quote I've already shared:

"The medieval element torp, however, must be seen in a context of the huge colonisation in northern Europe during the high Middle Ages, within a new ‘feudal’ agrarian system with a ‘manor’ and dependent tenant farms within an estate. In Germany these tenant farms often had the name dorf (< þorp), and the word for such a dependent farm was spread with the new colonising strategy to Scandinavia. Early on, the element torp must have developed into a meaning of secondary farm, a farm detached from a hamlet etc., hence not always denoting a tenant farm within an estate."

Edited November 8 by Genava55

[Genava55]

17 hours ago, Thalatta said:

There are many other things worth checking from that source about thorps which directly imply them being more than a single farmstead, for example this quote explicitly stating that that would just be the case for quite derelict ones:

Quote

Danelaw thorps first recorded before 1500 (264 out of 576) are now completely lost or are represented by little more than a single farmstead

One has to bear in mind that certain thorp characteristics don't necessarily apply to þurpą, being a cognate many centuries after the fact. All this focuses only on supporting the EE translation being a small settlement.

I do not think that an isolated example, removed from its immediate context, provides a sufficient basis for attributing to the authors a conclusion they do not themselves formulate. Other passages in the book show that their use of thorp does not necessarily presuppose a settlement composed of several houses or farms.

A particularly relevant example occurs in their discussion of minor place-names in Leicestershire, where they observe that "at least one was a farmstead in the thirteenth century". The reference is to Thorp on le Toftis, identified in a source from 1278 as "a lost farmstead described in 1278 as a tillage".

This demonstrates that, within the authors' own analysis, a place called a thorp could correspond to a farmstead or agricultural holding. It does not establish that every thorp had this form, but it does rule out the claim that the term necessarily referred to a group of several houses or farms.

Upthrop provides a comparable example. The authors treat this throp as a distinct unit in a landscape containing several dispersed settlements. Drawing on the charter of 869, they describe it as a 'little estate' incorporated into a larger estate complex. That description is entirely compatible with a small dependent agricultural establishment and does not, by itself, imply a village composed of several households.

This is consistent with their wider discussion of dependency. Some thorps recorded in the Domesday Book were classified as berewicks, meaning outlying farms or estates attached to larger centres. In another document, the dependent estates or appendices of Peterborough were translated as "þa þorpes". The category could therefore encompass subordinate agricultural or estate units, rather than only nucleated communities.

The same caution applies to the linguistic evidence. The surviving record of the Old English word is extremely limited, and the authors explicitly acknowledge how little is known about its early form and usage. Nothing in their analysis establishes that thorp originally had to designate a settlement containing several households. Consequently, the possibility that it could initially refer to a single farm, agricultural establishment, estate unit, or even a building at the centre of such a holding is not excluded. The last possibility is not directly demonstrated by the evidence, but neither is it contradicted by the authors' argument.

It is also important not to confuse the possible original meaning of the word with the particular historical form adopted by many medieval thorps. The authors' main concern is not the meaning of thorp at the beginning of the Anglo-Saxon period. Their analysis focuses primarily on the profound reorganisation of settlement and agriculture from the ninth century onwards, especially the spread of open-field farming and the movement from dispersed occupation towards more nucleated forms.

This is the context of the 'thorp moment'. The term describes an intermediate stage in the reorganisation of the medieval countryside, when small dependent units or satellite settlements helped bridge the gap between earlier dispersed farmsteads and later, more concentrated village territories. As the article puts it, 'Þorps and þrops represent an intermediate stage in the nucleation process, as farming communities sought the best way to run new field systems from the developing centres of population.'

The regular and sometimes nucleated layouts identified by the authors therefore belong to this medieval process of transformation. They do not prove that thorp had always meant 'hamlet' or 'village', still less that this was its only possible meaning in early Old English. A term that may originally have designated a farm, estate unit or agricultural establishment could later have been applied more specifically to the small dependent settlements created or reorganised during the thorp moment. Projecting that later medieval morphology backwards would turn the authors' historical hypothesis into a timeless lexical definition that they themselves never formulate.

Edited November 8 by Genava55

[Thalatta]

@Genava55 your first post is clear and I agree with most of it, except what you seem to actually stand for, which seems confirmed in your second post. Yes, "settlement" can be ambiguous, but many times it's clear what's being said from context, or even explicitly explained. When that book says "small settlements", it's clear that they refer to a few dwellings and not to the size of a single one. You argue against thorps usually "being more than a single farmstead" by referring to my quote as an "isolated example", but use a couple of isolated examples to say there's no "sufficient basis for attributing to the authors a conclusion they do not themselves formulate", when they actually conclude:

Quote

Briefly, let us restate some of the conclusions already reached in our analysis of thorp places. First, linguistic and archaeological indicators seem to agree in suggesting that thorps began as small, relatively insignificant settlements. In their earliest phases they appear to adopt simple, compact, linear plans, with a small number (rarely more than a dozen) of regularly sized building plots arranged along a single village street.

Which has been my point all along.

 

I wouldn't say an "unambiguous synonym for 'village'" is intended to be attached by current sources. As I've said many times, this association doesn't have to refer to a full-blown village, that might well be an exaggeration, it's just a matter of resemblance. When "hamlet or village" is used, what is meant is that there are multiple dwellings (or farmsteads, if you will), and not just a single one, and after all, all this kind of started when you said:

On 11/02/2026 at 12:02 AM, Genava55 said:

A village is a collection of dwellings; a farmstead is a single dwelling.

Even when that might be an exaggeration and "hamlet is a very few dwellings" should be put in-between. The central issue is clear and I've mentioned it many times: are we dealing with a single farmstead or a group of them? I think it's clear most of the current evidence points to the group, and why it's no big issue that the current EE translation uses "hamlet or village", not in the spirit of saying thorps were proper villages, but that they were "small settlements" made up from a "collection of dwellings". Maybe Ekwall (who, again, died over 60 years ago) wanted to call all that just "farm", possibly "hamlet", but not "village", but taking that old sentence literally is misleading, considering what the current evidence tries to imply.

 

And still, all that is only half of the argument. You still have not commented on Celtic cognates of haimaz having no relation with any kind of settlement, while þurpą cognates are consistently more settlement-like, showing it's doubtful to think haimaz mainly referred to some kind of settlement early on, while þurpą likely did. I still think you got too focused on Gothic and later (we are still talking about OE thorps after all) instead of the bigger picture provided by both branches and explaining why the accepted reconstructions are what they are.

Edited November 8 by Thalatta

[Genava55]

On 08/11/2026 at 4:02 PM, Thalatta said:

When that book says "small settlements", it's clear that they refer to a few dwellings and not to the size of a single one. You argue against thorps usually "being more than a single farmstead" by referring to my quote as an "isolated example", but use a couple of isolated examples to say there's no "sufficient basis for attributing to the authors a conclusion they do not themselves formulate", when they actually conclude:

Quote

Briefly, let us restate some of the conclusions already reached in our analysis of thorp places. First, linguistic and archaeological indicators seem to agree in suggesting that thorps began as small, relatively insignificant settlements. In their earliest phases they appear to adopt simple, compact, linear plans, with a small number (rarely more than a dozen) of regularly sized building plots arranged along a single village street.

I found this study fascinating. It is exactly the kind of research I enjoy, much like the studies of German and Swiss place names that I mentioned earlier. However, there is a reason why I have not placed too much weight on it in this discussion: the study is mainly concerned with the period from around 850 to 1250. The authors state this chronological focus several times. They examine that period because it was then that thorp and throp became highly productive in place names, during a major transformation of agriculture and rural settlement.

But that is not the central issue in our debate. We are trying to determine what these words may have meant much earlier, before the medieval period. Your claim is that Proto-Germanic þurpą already referred to a village or hamlet. That is what I am disputing. I challenged this interpretation using the Gothic evidence for two reasons. First, þaurp is never used to translate "village" in the Gothic Bible, despite the fact that villages are mentioned repeatedly. Second, its only occurrence refers to a field or agricultural estate.

Gothic is the only Germanic language for which we have substantial evidence from before the medieval period. You responded that Gothic may have undergone its own semantic innovation and lost an earlier village sense, while other Germanic languages later used cognates of þurpą for villages.

I do not deny the medieval evidence. At some point, these words clearly did come to designate villages. What I dispute is the assumption that this was their original meaning. My hypothesis is that the earliest meaning was closer to the one preserved in Gothic, and that the village sense developed during the later reorganisation of the agricultural landscape, when forms such as thorp and dorf became especially productive in place names.

I am restating this because our discussion seems to be drifting further into the Middle Ages, while the disputed question concerns a much earlier semantic stage.

The excerpt you quoted must also be understood within the chronological limits of the study. It comes from an archaeological analysis centered on the period from approximately 850 to 1250. When the authors refer to the "earliest phases" of thorps, they mean the earliest phases of the medieval settlement phenomenon they are reconstructing, the "thorp moment". To make this more explicit and clearer, I recommend looking at figure 7.3 on page 150 that I posted below. They are not making a statement about the original Proto-Germanic meaning of the word. "In their earliest phases" refers to the earliest archaeologically recoverable phases of the thorp settlements examined in the study. It does not refer to the earliest history of the word in Old English or the Germanic languages.

The compact and linear organisation described in the passage characterises the period when many thorps had become small villages or nucleated settlements. It cannot simply be projected backwards to prove that thorps had always denoted a settlement composed of several households.

The authors themselves do not claim that thorps appeared suddenly as fully formed villages. They consider the possible regrouping of earlier farmsteads and leave aspects of the preceding nucleation process unresolved. Their medieval reconstruction therefore cannot be treated as evidence that "village" or "hamlet" was already the word’s original meaning centuries earlier. This is notably what they are discussing at the pages 148, 149 and 150 of their book.

The passage actually describes a three-stage model: dispersed farmsteads, followed by small clusters of farmsteads, and only later by fully nucleated villages. The authors begin from an "archaeologically attested pre-open-field landscape of dispersed farmsteads" and question the conventional assumption that "full dispersion gave way to full nucleation" without any intermediate stage. Their alternative is explicit: "Might a first stage towards full nucleation not have been the grouping of individual farmsteads into a number of small hamlet clusters?"

In this model, thorps are precisely that intermediate stage. They are not necessarily settlements inserted into a landscape that had already been reorganised. Instead, the authors ask whether they "played [a role] in that reorganisation itself". A thorp could therefore emerge when previously dispersed farmsteads were brought together into a small secondary cluster associated with the laying out of new fields.

This helps clarify what the authors mean elsewhere when they describe the "earliest phases" of thorps as compact settlements with several building plots. They are describing the earliest archaeologically recognisable phase of the thorp moment, after individual farmsteads had begun to be grouped together. They are not claiming that the word thorp had always denoted such a cluster, nor that several buildings were necessarily part of its original meaning.

The authors are also careful to call the resulting settlement "a secondary or subsidiary or outlying hamlet," rather than a primary village. Their model is therefore not simply "thorp means village". It is a historical process:

Dispersed farmsteads → Small subsidiary clusters → Fuller village nucleation.

This passage consequently supports the idea that the characteristic medieval thorp could have resulted from the reorganisation of earlier individual farms. It does not allow the later clustered form to be projected backwards as the original and permanent meaning of the word. It is truly a new meaning that is emerging along with a new phenomenon.

This distinction is especially important because the authors themselves question whether the known English throp-names really belong to the earliest Anglo-Saxon centuries. They write that "it is not as clear to us as it was to Smith that the known place-names in þrop date from the earliest centuries of Anglo-Saxon settlement," and they describe Smith’s proposed pre-eighth-century chronology as "coherent, but […] far from conclusive." They also observe that þrop-names "tend to be recorded late" and lack the archaic or obscure first elements found in name-types believed to belong to the earlier Anglo-Saxon period.

They propose that thorps and throps played a part in the agricultural transformations of approximately 850 to 1250, involving open-field farming and increasing settlement nucleation. The "thorp moment" is presented as an "intermediary stage of nucleation". The compact plans and multiple plots described in the disputed passage belong to that medieval process, not necessarily to the original lexical history of the word. We cannot say that the definition of thorp between 850 and 1250 was the same much earlier. On the contrary, the authors dispute the existence of thorps before the 9th century. Therefore, you are misrepresenting their argument by twisting their conclusion to project it onto a different chronological period.

Furthermore, the example of a compact cluster of farmsteads should not be treated as the only possible model. It is mostly an ideal case. Elsewhere in the study, the authors discuss places called thorps that were represented by individual farmsteads or dependent agricultural holdings. This is why I think their use of the broad term settlement should be understood flexibly: it does not, by itself, specify a minimum number of houses, farms or households.

A few quotes from their book and their article I find appropriate to highlight the previous points:

Spoiler

"We therefore reject as unlikely earlier attempts to differentiate sharply the throps of English England [...] from the thorps of the Danelaw, and we suspect that both are likely to denote much the same sorts of settlement on the ground, established during a single phase [...] of settlement development."

"Thorps were born out of these transformations of farming practices and settlement patterns from the late ninth century onwards. But thorps, both as places and as communities, were not simply witnesses to these changes in the landscape: rather, they were active participants in them."

"The general implications of all these statistics are clear: thorps are much more likely to appear – if they survive at all – as small settlements or single farms than as parishes or villages"

"Subtle differences were also noted in their meanings. OE throp was noted in early documents as glossing terms such as tun, compitum (a cross-roads), fundus (estate) and villa (farm), and in one instance thing-stow (a place of assembly). Given these alternatives, early scholars proposed its translation as 'village' or 'hamlet', though medieval usage suggests that in names of late origin it was generally used of some smaller form of settlement."

"The word appears in five Anglo-Saxon glossaries, although four of them ultimately derive the relevant item from a single source, which rather reduces the witness-count. The exception appears the most straightforward instance. A list copied in the first half of the eleventh century contains the gloss ‘fundus, þrop’, where fundus is a term defined by Lewis and Short’s Latin Dictionary as ‘a piece of land, a farm, estate’. This clearly sits within the range of senses we might expect for throp/thorp. The other group of glosses seems more challenging. Here throp appears in association with Lat compitum, which primarily means ‘cross-roads’. In one instance it is additionally combined with OE thing-stow ‘meeting-place’. These senses are rather different from what is generally suggested by Germanic cognates and later English evidence, but they have been admitted into the dictionary definitions of throp, have been sought in some place-names and have coloured discussions of the Germanic term’s ultimate etymology."

"This book focuses on the thorps and throps of England. But, of course, these forms of place-name have a far wider geographic distribution. Names in -torp can be found today in great numbers in Denmark, where over 3,000 of an estimated 7,000 still survive, and they occur with frequency in Sweden, Norway and the Netherlands. They are present in lesser numbers in northern Germany, Finland (where around 800 have been recorded), Iceland and France. The element -torp lies behind Danish settlement names in -rup, -strup, -trup and -drup; in Germany it appears as -dorp, -dorf and -dörp; and in the Netherlands it is the root of Terpen names. Consequently, the study of thorp names is an international concern. It is a lively field of enquiry, warranting major conferences devoted exclusively to the subject and producing an ever-growing body of literature. What European studies have shown is that thorps tend to be small, often dependent, low-status and outlying places. This is true across their whole distribution. Beyond this, however, they appear to share little in common. It has been suggested, for example, that thorps in Denmark have their origins in a period of settlement expansion and landscape colonisation beginning in the last quarter of the tenth century and concluding before 1100."

"Additional early senses were offered. Gothic thaurp ‘a field’ was identified as a cognate, while Primitive Germanic *thurpa- was associated with Latin turba ‘a tumult, a mob, a crowd’. The latter was linked to a meaning ‘village’, and supported by the fact that in Old High German the term dorf could mean ‘a gathering of people’. Smith identified Germany as the early centre of a place-name element already well established by the fifth and sixth centuries. From here it spread into southern Denmark, where it had become common by the Viking period. It continued its northerly advance into Sweden, and ultimately, but in a more limited way, to Norway and Iceland. Smith argued that it entered England with an early wave of Anglo-Saxon settlers coming from the Germanic homelands."

"There is general agreement that the þorps of the Danelaw were originally diminutive in size and dependent on a larger settlement nearby: A. H. Smith defined the term as ‘a secondary settlement, a dependent outlying farmstead or hamlet’, observing that this is closely equivalent to the sense of þorp in early Denmark (Smith 1956, II: 205–209). Later commentators have been happy broadly to agree. And scholars have also recognised that there were good reasons to interpret the þrop names outside the Danelaw in much the same way: Smith glossed this element as ‘a hamlet, an outlying farm’, and indicated in his discussion that there were clear signs of dependency here as well (ibid.: 214–215)."

"Such an account implies that the distribution of the þrops may be governed by factors other than the circumstances of the earliest Anglo-Saxon settlements, and it is clear that Fellows-Jensen had serious doubts about Smith’s chronology. She went on to note that the þrops are often rather late-recorded, later than the Danelaw þorps, suggesting that the English names ‘may be younger as a class’; she has even floated the idea that some of the þrop-names may have arisen under the influence of Danish þorp as a loan-word (Fellows-Jensen 2003: 49)."

"There seem to us also good reasons to doubt Smith’s chronology. Þrop-names do tend to be recorded late, even though – as Fellows-Jensen observed – they are found in areas where early documentation is relatively abundant. Þrop is not combined with the sort of archaic or obscure first elements that characterise place-names containing other elements, like hām ‘homestead’, believed to belong to the earlier Anglo-Saxon centuries. And we, like Fellows-Jensen, are struck by that continuous, coherent looking, geographical distribution, and the great apparent similarity in usage between þorps and þrops either side of the Danelaw line. It may be possible that the settlements named in this way were created on similar lines centuries apart."

 

On 07/11/2026 at 5:42 PM, Thalatta said:

And you saw but omitted all that, because that's followed by the sentences you quoted from Isidore of Seville regarding what a compitum is, and then even said:

Isidore of Seville writings are from the 6th century. The Épinal-Erfurt Glossary is from the 7th century. So once again, this is a chronological period much earlier than that observed archaeologically by the authors of the book we have just discussed. The authors do not address glosses directly in their study but in an appendix, and they express legitimate doubts. However, a large part of their criticism is directed at the concept of a crossroads associated with the word conpetum. Personally, I agree with them and I absolutely do not support this interpretation, which derives more from classical Latin than from the Middle Ages. Isidore of Seville, on the other hand, emphasizes the agricultural nature by saying it is a gathering of peasants.

I really think it's important to gather all the early elements related to the cognates of þurpą. We see that the word thorp can be associated with the notion of gathering, both with its comparison to concetum and thing-stow, but also by looking at other cognates in German and Old Norse that can also be used in a context of gathering.

This is why I said that 'þrop' referred to an estate acting also as a gathering place for peasants who collectively managed land, and that there was a semantic shift in its meaning to refer first to the entire community and then to a village. In my opinion, the comparison with tuun is more logical in this sense because this word has also undergone a similar semantic shift, with the notion of a physically delimited space to then designate the notion of property, of domain and then the buildings which are contained in the domain. I agree with the authors when they say that these are "settlements" but I insist that this could derive from a semantic shift, first starting with the idea of bringing together peasants initially, then bringing together farms and agricultural holdings in a second stage. The authors closed the debate too quickly by assuming that "tuun" must refer to a settlement and that, therefore, "throp" necessarily referred to the same thing even then. I think they should have explored this question more deeply than it might seem, beyond a simple appendix, as I believe it would have lent even more weight to their concept of community-based agricultural land management. Looking at the problem from the perspective of a delimited space dedicated to agriculture can help to unify the different definitions given in the different glossaries.

On 08/11/2026 at 4:02 PM, Thalatta said:

And still, all that is only half of the argument.

I will respond to the rest of the arguments in due course. I want to respond properly to your arguments.

Edited Tuesday at 00:30 by Genava55
move the quotes in a spoiler section for readability

[Thalatta]

There's the discussion regarding þurpą reconstruction, and there's the later discussion regarding the attested cognates. Regarding the latter, you have been trying to argue that one can see the earliest attested cognates refer to a farmstead, or according to you, a single dwelling, first with OHG, but then I've shown that OE is attested earlier. Concerning that latest discussion, "original meaning" can only refer to the time of the EE glossary. Then, although "the study is mainly concerned with the period from around 850 to 1250", it's not correct to say that "they are not claiming that the word thorp had always denoted such a cluster, nor that several buildings were necessarily part of its original meaning", since I've shown they do claim this when they stated that compitum in the EE must have also been interpreted as a "small rural settlement" (conclusion I arrived at independently, in part considering they would emerge right there) by "unsophisticated people". But you know this, since later you dismiss this evidence by saying that

Quote

The authors closed the debate too quickly by assuming that "tuun" must refer to a settlement and that, therefore, "throp" necessarily referred to the same thing even then. I think they should have explored this question more deeply than it might seem, beyond a simple appendix

It's not reasonable to suppose that more exploration would have led to the conclusion that "thorp" was actually a single dwelling at the time of the EE glossary, which was your claim. That would be basically the opposite of what they are already claiming, which aligns with the current EE translation (published a few years later), as I've already explained. But maybe focusing on þurpą in what follows will clarify this even further.
 

Concerning þurpą reconstruction (the other interpretation of "original meaning"), my claim is not exactly that "Proto-Germanic þurpą already referred to a village or hamlet", that's semantically charged, but that þurpą probably didn't restrict itself to a single farmstead. Then I guess you could call this collection of farmsteads a settlement, a hamlet, or small village, but none of that is my original claim (just approximate terms to emphasise a contraposition with your single dwelling claim). My claim is based on that (new relevant information is added in the last point):

-The Germanic cognates you tried to use to further your claim actually support mine, as demonstrated from your sources. The rare unique instance in Gothic cannot be reliably used, and could have treated a collection of farmsteads as "farmland", like Ekwall seems to do.

-The Celtic cognates also refer to a settlement larger than a single farmstead, further reinforcing the view that, if that was the actual meaning in Gothic, it was indeed the exception. This is way more parsimonious than thinking that the "earliest meaning was closer to the one preserved in Gothic" and everything else changed.

-The Proto-Germanic reconstruction is often associated with "crowd" and "settlement". While you pointed out that the latter is ambiguous, Kroonen's student states, when questioning if Latin trabs is also a cognate, that "while all the other comparanda refer to dwellings, settlements, and communities, Lat. trabs sometimes refers to part of a building". Through his thesis he consistently uses the word "settlement" to describe multiple dwellings, and small "communities" is exactly what I've been talking about (conclusion I arrived at independently, in part considering the association with "crowd"). Notice how, even when listing in the previous page that in Gothic it meant "farmland", absolutely nothing is said about that for Proto-Germanic.

 

Regarding the quotes from the book towards the end, I know them, they either align with my claim or don't go against it, and certainly don't support a single dwelling interpretation. Many specific characteristics regarding thorps is why I said to bear in mind they "don't necessarily apply to þurpą, being a cognate many centuries after the fact". The relevant claims in question are the ones I just restated in this post.

Edited Sunday at 22:58 by Thalatta

[Genava55]

On 07/11/2026 at 5:42 PM, Thalatta said:

Haimaz: at least the meanings home, homely and homestead are justified (by your own sources). Gothic has a stronger case here, although it translates haimaz cognates as country at least once, inverts wīhsą cognates urbanistic order, and all this is centuries after the fact. You tried to argue for it meaning settlement using Old High German, but this is ten centuries after the fact. Haimaz cognates ended up meaning settlement as in "someone's home", but this was not its original meaning, not only because the homely meaning is ever present and ambiguating in the Germanic branch, but because it's astoundingly dominant in the Celtic branch, showing that by the time PIE split into these two branches there was probably no settlement meaning yet, and the lack of any horizontal transfer hints to a rather late than early development, but it cannot be said at which point between Proto-Germanic and Gothic this meaning appeared. All this shows that haimaz early on meaning mainly settlement seems unlikely. Its meaning refers more often to a home.

I believe I have defended my position several times regarding my suggestion. Cognates derived from “haimaz” that mean “village” or “settlement” have the advantage of being supported by the Gothic language, which is the attested language chronologically closest to Proto-Germanic. There are also clear examples of their use to mean “villages” and “hamlets” in Old English, Old High German and Old Frankish. Finally, place names ending in -heim, -hem, -ham, etc., are among the oldest Germanic place names, representing one of the earliest chronological layers of naming.

But since my arguments have already been lost amid the recent debates, I will present a fresh set of arguments. To do so, I will begin by quoting at length from a scholar I already mentioned, Stefan Brink. In 1995, he wrote a chapter titled “Home: The Term and the Concept from a Linguistic and Settlement-Historical Viewpoint.” While it is not an exhaustive compilation of all the evidence, since it focuses on his extensive knowledge of the Scandinavian and Anglo-Saxon worlds, it serves as a very solid foundation from which to begin.

1. A collective meaning is part of the inherited Germanic evidence

The modern meaning of home should not be projected backwards as though it exhausted the historical semantic range of “haimaz”. As Brink points out, ancient related Germanic words could refer not only to an individual dwelling or a farm, but also to an estate, a group of dwellings, or a village. This makes “haimaz” a plausible term for a primitive residential community rather than simply a private home. In this introductory excerpt from his article, he provides a few examples and makes it clear that the meaning of a collective dwelling is entirely valid:

Spoiler

 “One can only go to oneself and ask for the meaning of home. The semantic core is probably shared with most other people in the same cultural context, however the semantic picture will probably fade out at the edges. What probably comes to mind, is one's living-house with various “necessities” in the vicinity. A more collective valid meaning we can look up in lexica. If we however for some reason want to use home as an unambiguous term in research, we of course have to give it a precise definition. This is the synchronic life of the word home, its use today. What about the use of the term and the concept home in earlier days, hence what about the diachronic aspect? The word home has equivalent cognates in most Germanic languages (here the English form home is used for all Germanic related words). The forms and lexical meanings in older Germanic languages are e.g. OEng. hām n. (collection of dwellings, village, estate, house etc.), OHGerman heima m. (home, world), OSaxon hem (home) and OScandinavian heimr m. (dwelling, home, world). The meaning home, or one's farm, goes far back in history for the Scandinavian word heimr, hem.”

2. Gothic and Lithuanian provide direct and ancient evidence of the meaning of the word “village”

The Gothic evidence is especially important because it is considerably earlier than most of the surviving West and North Germanic material. Wulfila’s use of haims to translate Greek kōmē, “village” or “hamlet”, demonstrates that the collective settlement meaning was already established in a fourth-century Germanic language. The opposition between plural haimos, “countryside”, and polis, “town”, also places the word within a clearly rural settlement system. Stefan Brink considers these meanings derived from Gothic to be consistent and convincing. He even notes the similarities with two loanwords found in Lithuanian that have virtually the same meaning as in Gothic, conveying both the sense of a village and that of the countryside:

Spoiler

“The word is also found in the lost Gothic language in the form haims. Gothic is handed down to us with the Bible as a filtering-medium, which must be kept in mind. Here, Wulfila the translator in the 4th century A.D., used haims for the Greek word κώμη (village, hamlet), but even more interestingly he has used the plur. haimos for Greek ἀγροί (country, country-side) to oppose πόλις (town). Of course, this early observation will be of importance in the following discussion. It is also worth noticing the two adjectives Gothic *af-haim(ei)s (absent, away from home) and *ana-haim(ei)s (present, at home) (Feist 1939, 233 f.; Lehmann 1986, 170). The Proto-Germanic form is *χaima- (-i-), which was most probably borrowed into the Lithuanian kiemas (village, farm) etc., and káima(s) (village, land or country as opposed to town) (Fraenkel 1962, 251). The ultimate relations of this Proto-Germanic base χaim- are somewhat disputed, however most probable, and in agreement with most etymologians, it is to be understood as an abstract formation with an m-suffix to an IE. root kei- (to lie), as in Greek κεῖμαι (lies), κοιμάω (brings to rest), cf. κοίτη (lair, camp). It is interesting to note that the same formation is found in e.g., Latvian saime (family, household servants) and OSlavic sēmija (household servants; slaves). (Pokorny 1959, 539 f.; Kluge 1989, 301; Hellquist 1948, 346.)”

3. The toponymic evidence confirms that the word could designate an inhabited community

The place-name evidence independently supports the collective use found in Gothic and Old English. Names in -ham are interpreted not only as references to farms or estates, but also to hamlets, villages and the territories occupied by particular communities. Formations such as Amersham and Gillingham show that the element could identify the settlement or district associated with a person and their followers, rather than merely that person’s individual home:

Spoiler

“If we turn to the toponymical evidence, several interesting observations of great value in the discussion of the word home are to be found. The equivalants to home are found in many old place-names on centrally located settlements all over the North Germanic language area. In Germany the place-name element in question is-heim, whereas in England it is-ham. This latter element is rather problematic, since place-names in-ham have different origins. Besides the old OEnglish-ham with assumed meanings as (village, estate, farm), we have some other words that are found amongst the-ham names, namely OEnglish hamm (enclosure), hamm (river-meadow), hemm (edge, border) etc. (Sandred 1976). In England the-ham names show an uneven distribution, e.g., East Anglia have many names, whereas the western Midlands have few. Normally the first element in-ham names are a man's name, Amersham (Ealhmund's village), however in some cases the first element is a topographical word, Burnham (village by the stream). Of special interest are the-ingaham names (village of X’s followers), thus containing an inhabitant-name in-inga and the name of some leading man of this people, e.g. Gillingham (the homestead of the followers of Gylla), Wokeingham (the homestead of the followers of Wocca). These names are related to the-ingas names, where the element-ingas (the followers of X) also contains a man's name as the first element, e.g. Reading (the followers of Read) and Hasting (the followers of Hæsta). Of course, these latter names do not have to denote a habitation etc., it is possible to interpret these with a more areal meaning, i.e. settlement-district (Dodgson 1973; Cox 1973; Gelling 1978, 112 ff.).”

4. Specialists generally reconstruct an original meaning centred on settlement

The Scandinavian material is particularly valuable because it preserves both the territorial and residential dimensions of the word. Despite continued debate over the precise scale of the referent, Brink observes that most scholars regard “settlement” as the original meaning of -he(i)m. The underlying concept is therefore not simply a house, but the place where a person or community has settled:

Spoiler

 “If we turn to the-hem, -heim names in the Nordic countries, we find an interesting difference compared to the British and Continental names in-ham and-heim. The first obvious thing is that the Scandinavian names never seem to contain a person's name as the first element. Instead we find topographical appellatives. (The same opposition we find regarding the Scandinavian-inga names!) A valid starting-point for the discussion of the Scandinavian-hem names, may be to contrast e.g. the names Trondheim, in Norway, with Sähem, one of several Swedish names. The former means (the settlement-district of the trønder [in Trøndelag]), while the latter means (the farm by the lake). This opposition in meaning, an area contra a habitation, has led to a vivid discussion amongst scholars what is the original meaning of-he(i)m. Most scholars have come to the conclusion that the latter meaning, (a farm, a settlement), represents the original meaning for this element, and one has especially used the meaning of the Greek κεῖμαι (lie) as a tool to come up with an older meaning (place where one has settled; settlement).”

5. The settlement use is demonstrably ancient

The collective or territorial interpretation cannot be dismissed as a late medieval development. Brink identifies a group of very old Scandinavian -he(i)m names that were probably formed around the beginning of the Common Era, including one securely dated to approximately AD 200. This places the use of haimaz and its descendants as settlement names within the chronological horizon relevant to early Germanic societies:

Spoiler

 “If we look at the actual material, we find that among Norway's ca. 1,000 heim-names, Sweden's ca. 300 hem-names and Denmark's ca. 200 hem-names, a large part are somewhat stereotypical names, e.g. Solhem (sun—), Nyhem (new—), Opphem (upper—), Sähem (lake—), Askhem (ashwood—), Grythem (stone—), etc. Many of these names must be considered to be fairly young, from the Late Iron Age. However there are some really old-he(i)m names as well, spread nearly all over the Iron Age settlement districts in Scandinavia. These names have probably been coined already in the centuries around the birth of Christ. With a combined onomastical, archaeological, and palaeo-ecological dating, I have been able to give the hem-name Vattjom in the province of Medelpad, northern Sweden, a precise dating to as early as ca 200 A.D., probably the most “secure” dating such an old place-name has in the whole of Scandinavia (Brink 1984, 45 ff.).”

6. The word could include both the habitation and its surrounding resources

Brink ultimately argues that the traditional opposition between “habitation” and “settlement district” is too rigid. His proposed meaning, “settlement or settlements with an area of resources”, accounts for both an individual farm and a wider community or territory. This is highly relevant to an early Germanic “village”, which need not have consisted of tightly clustered buildings but could instead have been a socially coherent settlement together with its fields, pastures, woodland and other resources, exactly what we have in the game:

Spoiler

 “In the discussions so far, one has concentrated too much on the question whether-he(i)m denoted (a habitation) or (a settlement-district). This has led to an academical hair-splitting. Obviously there is nothing etymologically that forces us to presuppose some kind of habitation (hence actual buildings) for the meaning of home, haim, hem, heim, hjem etc. Therefore, it is tempting to assume that a vital semantical component for the-he(i)m names has been (settlement with area of resources). With this explanation, there is nothing problematic to have on the one hand names like Trondheim (the province of the trønder) or Böhmen (the district occupied by the Boii), and on the other Sähem (the farm by the lake) or Stenhem (the farm by the stone); and actually this interpretation also finds support when we look at the appellatives, such as the Gothic plural haimos (country) and OWScandinavian heimr (dwelling, home; world) (Brink 1991).”

7. The semantic core is an inhabited territory, not merely a building

Brink’s conclusion brings the different strands of evidence together. The essential meaning of haimaz and its cognates was the place and surrounding area where a person or group lived. Depending on scale and context, this could be a farm, settlement, village, community territory, or settlement district. This semantic flexibility does not weaken haimaz as a phase name. On the contrary, it makes it particularly well suited to an early form of village whose identity rested on shared habitation, territory and belonging rather than on urban density.

Spoiler

 “To sum up, regarding the place-name element -ham, -hem, -heim, what must have been of interest for the name-givers have been the area — small or large — where some people have lived. This explains names such as German Böhmen (Bohemia, Latin Boioh(a)emum) (the land of the Boii), English Gillingham (the district of the followers of Gylla), Norwegian Trondheim (the district of the trønder), Swedish Grythem (the farm on the stony terrain [and its possessions]). And if we turn back to the etymological considerations of the appellatives, this areal component seems to be vital. Thus one could end this discussion by stating that some kind of semantical core, regarding the word home and its cognates, is the area and place where you live, a place with its surroundings that you normally take an affection to and have a very special relationship to, throughout your lifetime.“

This concludes the excerpts from the chapter written by Stefan Brink. I will also highlight some other points that have already been mentioned.

Here an excerpt of the description from the Deutsches Ortsnamenbuch (2017) about -heim place-names:

Spoiler

-heim. Germ. *haima- ‘Home of a tribe’ (Heimat) ; in the individual Germanic languages with various stem formations and genders, e.g.: OHG heima Fem. ‘Home, homeland, residence’, late OHG/MHG heim Neutr. ‘Homeland, dwelling place, house’, OSax. hēm Neutr., MLG hēm(e) Fem. / hēm also Neutr., OFris. hām / hēm Masc. or Neutr., ON heimr Masc., OE hām Masc. ‘Village, estate’, Goth. haims Fem. ‘Village, small town’. The latter meaning likely applied from the beginning to -heim group settlements, although -heim originally occurred for individual settlements (farmsteads) as well.

The -heim names, like those ending in -ingen, show characteristics of great age. They likely occurred sporadically as designations as early as the early phase (around the birth of Christ) and then became common during the early land acquisition (frühn Landnahme) of the 3rd–5th centuries (perhaps also as a translation of the Latin villa). In contrast to the PN-orientation (personal name) of -ingen names, the defining factor here is possession (‘Home / Estate of ...’). By the Merovingian period, the type was fully established and remained productive until the Middle Ages (MA), though varying by region.

I will also add two excerpts I already posted from the digital portal for place-name research in Switzerland:

Spoiler

The name-type -heim, which is widespread across a large part of the German-speaking area, is only sparsely attested as an ancient form in German-speaking Switzerland. Usually, ancient -heim is combined with a personal name [PN] as the defining element (Bestimmungswort), thereby naming a hamlet or a village and its leader.

[…]

The ‑heim and ‑dorf names in northwestern Switzerland are not entirely easy to interpret. ‑heim names are usually assigned to the oldest naming layer. In our region they occur only twice (Arlesheim and Riehen) and are generally regarded as outliers of the numerous Alsatian ‑heim names. ‑dorf names are likewise rare here; apart from Rodersdorf, all of them lie near the Hauenstein route. Niederdorf and Oberdorf are clearly secondary. They were introduced in 1285, after a massive landslide, to designate the remaining parts of the village of Onoltzwil. Frenkendorf, Füllinsdorf, and Arisdorf—since they lie in the middle of a zone with demonstrably unchanged Romance names—are hardly to be counted among the oldest Alemannic layer either, but should rather be attributed to the period of territorial expansion in the 7th and 8th centuries. The remaining Alemannic settlement names, which all belong at the earliest to the period of Alemannic territorial expansion, are distributed fairly evenly across our region.

Another excerpt, this time from the German National Atlas:

Spoiler

 The oldest settlement names in Central Europe date back to pre‑Germanic times. They are of Celtic or Romance origin and appear above all in the western and southern Germanic regions of the Germania Romana. The earliest Germanic names occur somewhat more frequently; they often end, for example, in ‑lar or ‑mar. After the Migration Period, beginning in the 4th century, the first major wave of land appropriation took place, during which fertile and easily cultivable areas were settled. Typical place‑name endings from this period include ‑heim, ‑ingen, ‑stedt, and ‑stetten. In the Merovingian era (6th–8th century), the settlement area was expanded (early period of development). Settlement names ending in ‑dorf, ‑hausen, and ‑weiler generally date from this time. The greatest expansion of settlement in Central Europe occurred during the clearing periods beginning around the 8th century. The old settlement landscapes were largely populated; people then began to push into the low mountain ranges. At first, the edges of the mountains and the valleys were settled (e.g., ‑bach, ‑born, and ‑brunn). From about the 10th century onward, and throughout the High Middle Ages, the forest was increasingly pushed back and new settlements were established. Typical place‑name endings from this period indicate the clearing process, such as ‑roth, ‑rieth, ‑reut, ‑brand, ‑schwand, ‑hau, ‑schneid.

Taken together, these sources point to the same conclusion. Place names derived from haimaz belong to one of the oldest settlement-naming traditions of the Germanic world and are consistently associated with villages, hamlets, estates, and inhabited communities rather than isolated dwellings. In the Old High German sphere, -heim names are generally regarded as belonging to the earliest layers of settlement nomenclature, appearing at least during the Migration Period and possibly even earlier. The continuity of this evidence across Germany, Switzerland, and the wider Germanic-speaking world strongly supports the interpretation of haimaz as a term connected with settlement and community from a very early date.

Similarly, Richard Jones, David N. Parsons, and Paul Cullen take the same position regarding England: place names related to the word “ham” date back to the earliest period:

Spoiler

“Þrop-names do tend to be recorded late, even though – as Fellows-Jensen observed – they are found in areas where early documentation is relatively abundant. Þrop is not combined with the sort of archaic or obscure first elements that characterise place-names containing other elements, like hām ‘homestead’, believed to belong to the earlier Anglo-Saxon centuries.”

Albert Hugh Smith himself believed that place names derived from “ham” were part of the oldest layer of place names in England:

Spoiler

hām (hām dat.sg.) OE, 'a village, a village community, a manor, a homestead'. (1) The root meaning of hām and its many cognates is 'a safe dwelling' (v. NED s.v. home), and in the Germ languages it denoted 'a village' (cf. Goth haims) or 'a dwelling-place' (OSax hem, OHG heim, ON heimr) as well as more general ideas such as 'one's country' (OHG), 'the world' (ON). The meanings certainly established in OE are: (i) 'a dwelling-place, a house', as in Lindisfarne Gospels (John xiv. 2) In hus fadres mines hamas meniga sint, 'in my father's house are many mansions', where hamas glosses Lat mansiones and is equated in the WSax Gospel by eardungstowa 'dwelling-places'; v. also hām-stede, hām-tūn. (ii) 'a village, a collection of dwellings', as suggested by such contexts as those of the OEBede ii, 14 he rad betweoh his hamum oðþe be tunum for Bede's inter civitates sive villas, a hām being apparently something larger than a tūn, or of ASC 1000 and for-bærndon Tegntun and eac fela oðra godra hama, 'and they burned Tainton and also many other good villages'; this sense of 'village' is not found later than early ME (cf. NED). (iii) 'a manor, an estate', which is evidenced in glosses such as hame for Lat predia, to hyre agenum hame for ad prediolum suum (BTSuppl s.v. hām) and would seem to occur in compounds like biscopham 'episcopal manor' (Crawf 23, BCS 706), cyneham 'royal manor' (BCS 165, 1220). (iv) 'a household, a monastery', as in Bede's abbas in monasterio quod vocatur Peartaneu (iii, 11) rendered in the OEBede as abbud of Peortanea ðam ham and in the compound mynster-hām (Thorpe 482, Alfred's Laws 2); cf. F. E. Harmer, English Historical Documents 14. In p.ns. the precise meaning cannot usually be determined, but in the majority it probably denotes 'a village'. (2) The el. hām is found in most parts of the country but its frequency is greatest in the SE (K, Sr, Sx), in the Thames valley, in parts of EAngl (Nf, Sf, Ess, C), whilst further north it is fairly common in YE, Du and Nb. It is relatively uncommon or rare in the NW (Cu, We), the Midl (Lei, Wa, Nth, Bd, Hu, Hrt) and the WMidl and SW. This distribution would suggest that hām belongs to the earlier period of the English settlement and that it was becoming obsolete as a p.n. term as the settlement advanced towards the west. Other facts pointing to the same conclusion are: (i) most p.ns. containing this el. are DB manors and the el. is rarely found in minor names (except ME and later f.ns. like Homfeld Mx 199 Home Field W 434 where it is rather ModE home meaning '(some thing) near the farmstead or residence'); (ii) it does not enter into post-Conquest p.ns. combined with OFr pers.ns. as do bý and tūn; (iii) it occurs in compounds of which the other els. are apparently old (such as burna but not brōc), with pers.ns. thought to be very old, as well as ancient family names of the -ingas type (v. -ingahām); (iv) it is sometimes found with strong adjs., as in OE hēah-hām, Higham Bd, Db, Ess, K et passim, whereas with tūn the later weak adj. type is usual as in hēa-tūn, hēan-tūne (v. hēah).

Finally, there is also Old Frankish, which provides fairly strong evidence that words derived from haim often had a collective meaning. In addition to the fact that many linguists suspect this is the origin of the French word “hameau”, there is also ample evidence derived from an analysis of legal texts from the Frankish Empire. Maurits Gysseling’s analysis of the Germanic vocabulary preserved in the Lex Salica provides substantial evidence from the Frankish linguistic area. His reconstruction of haima- and his analyses of anði-haim-ja- and haima-þeuða- consistently support a collective sense equivalent to “village”. In several independent contexts, haima- clearly refers to an organized rural community rather than merely to a private household. This makes the collective meaning part of the Frankish evidence itself, not a later interpretation inferred solely from place names.

To summarize, the strength of the argument does not rest on a single piece of evidence, but on the convergence of multiple independent lines of evidence, all pointing in the same direction:

Spoiler

1. Gothic provides direct evidence: the attested Gothic word haims was used to translate Greek kōmē (“village, hamlet”), making the meaning “village” an explicit part of the earliest surviving Germanic evidence rather than a modern reconstruction.
2. Old English preserves the same collective sense: hām is repeatedly attested with meanings such as “village”, “collection of dwellings”, “estate”, and “settlement”, demonstrating that the word could denote an organized community rather than merely an individual house.
3. The Frankish evidence points in the same direction: linguistic and legal evidence from the Frankish world repeatedly supports a collective meaning connected with villages, rural communities, and organized settlements rather than simply a private household.
4. The wider Germanic evidence is remarkably consistent: cognates throughout the Germanic languages display semantic ranges extending from dwelling and farm to settlement, inhabited territory, community, district, and homeland, suggesting a concept broader than that of a single building.
5. Toponymic evidence independently confirms this interpretation: place-name elements derived from haimaz (-heim, -hem, -ham) are routinely interpreted by specialists as referring to villages, hamlets, settlement districts, inhabited territories, or communities.
6. Specialists generally reconstruct an original meaning centred on settlement: as Stefan Brink notes, the prevailing scholarly interpretation derives haimaz from the idea of a place where people have settled. The most widely accepted reconstruction is therefore closer to “settlement” than to “house.”
7. The semantic core unites people and territory: the evidence supports meanings ranging from farm and village to settlement district and inhabited landscape, all connected by the notion of a community established in and identified with a particular place and its resources.
8. The Anglo-Saxon evidence is chronologically significant: names in -ham belong to one of the oldest layers of English place-names and are generally associated with the earliest phases of Anglo-Saxon settlement.
9. The Continental Germanic evidence is equally ancient: names in -heim form one of the oldest strata of Old High German toponymy. Most scholars associate them with the Migration Period, while some place their origins even earlier, in the Roman Iron Age.
10. The Scandinavian evidence pushes the chronology back further still: -heim/-hem names belong to one of the oldest naming layers in Scandinavia, and at least one example has been independently dated to approximately AD 200, firmly within the Roman Iron Age.
11. The Baltic evidence reinforces the reconstruction: early Germanic loanwords in Baltic languages preserve meanings connected with villages, rural communities, and inhabited territories, providing external support for the collective interpretation of haimaz.

Taken together, the Gothic, Old English, Frankish, Baltic, Germanic, toponymic, and settlement-historical evidence converges on the same conclusion: haimaz was not simply a word used exclusively to refer to a house or a private residence. Across both language and place-name evidence, it consistently appears as a term that could denote a settlement, village, inhabited territory, or organized community. Moreover, the forms derived from it belong to some of the oldest naming traditions in the Germanic world, making haimaz not only linguistically defensible, but arguably one of the best-attested and most historically appropriate choices available.

Edited 3 hours ago by Genava55
dealing with this awful forum messing everything

[Thalatta]

The discussion was never about its latter attested haimaz cognates not meaning some kind of settlement (are you getting confused with the þurpa discussion?), but if that was the case in Proto-Germanic times. That's why I agree with the quotes you bring up, but they are completely irrelevant to the point in question. By "early" and "late" I obviously meant in reference to the Germanic-Celtic split, bringing up the "late medieval" period is quite the misrepresentation.

Your source (Brink) does address what you keep ignoring regarding Celtic by saying "in Old Celtic languages we find a related formation that demonstrates another interesting theme, namely the meaning 'love'", and as you quote, concludes his chapter with:

Quote

Thus one could end this discussion by stating that some kind of semantical core, regarding the word home and its cognates, is the area and place where you live, a place with its surroundings that you normally take an affection to and have a very special relationship to, throughout your lifetime.

This is exactly my point (yet again you are brushing off the actual conclusions of your own sources). He refers to one's home (one's "area and place"), this doesn't necessarily point to a type of settlement, and at no point he claims that in Proto-Germanic times it referred to anything outside one's own (as seems to happen in Gothic and later, but it's completely different in Proto-Celtic and later).

 

3 hours ago, Genava55 said:

This is highly relevant to an early Germanic “village”, which need not have consisted of tightly clustered buildings but could instead have been a socially coherent settlement together with its fields, pastures, woodland and other resources, exactly what we have in the game

I completely agree, since this is my exact argument regarding þurpa.

 

3 hours ago, Genava55 said:

Depending on scale and context, this could be a farm, settlement, village, community territory, or settlement district.

I'm glad you brought up this point, since it's the one I was going to follow up with. Two characteristics are important to name a Phase: its "multiple dwelling" quality, and its development stage. Regarding Proto-Germanic haimaz, your source doesn't claim anything beyond that it referred to one's home, even if that could have indirectly referred (only to) one's own settlement, and we both agree it can be of any size. Regarding Proto-Germanic þurpa, I've already proven its "multiple dwelling" quality using your own sources, and we both agree that it's small.

 

3 hours ago, Genava55 said:

This semantic flexibility does not weaken haimaz as a phase name. On the contrary, it makes it particularly well suited to an early form of village whose identity rested on shared habitation, territory and belonging rather than on urban density.

Simply put: no. Urban development is the actual point of Phase names, and you know it, that's why you are pre-emptively trying to dismiss this obvious conclusion supporting what I say (as you've tried to do before). You have reached, in good part by yourself, the conclusion that þurpa is more appropriate than haimaz for Phase I, according to each of both important characteristics mentioned (being multiple dwellings, but a few).