Issue 370: Man- based class names
Posted by Robert Sanderson on 4/4/2018
Dear all,
Given that the names of the classes (and relationships) are not fixed, it would be wonderful if we could rename the classes that have “Man-“ in them to be more gender neutral. In particular, this would result in name changes for:
E22 Man-Made Object
E24 Physical Man-Made Thing
E25 Man-Made Feature
E71 Man-Made Thing
The smallest possible change would be simply to drop “Man-“ and leave the “Made”, thus E22 Made Object, E25 Made Feature and E24 Physical Made Thing.
Alternatively, “Artificial” is synonymous with Man-Made, or to make it into a noun, an “E22 Artifact”, compared with an E25 Artificial Feature and an E71 Artificial Thing.
“Produced” would fit in well with Production, but sounds less obvious to me, at least.
Other (IMO worse) options could include “Synthetic” (which would be strange for objects made from all natural materials), “Manufactured” (implies industrial scale with machinery), “Constructed” or “Built” (implies a large object such as a building or road).
Many thanks for your thoughts,
posted by Thanasis Velios on 4/4/2018
From my short experience of explaining/teaching the CRM in English to
conservation students, the issue of gender neutral labels has come up a
few times. I think it should be discussed as an issue. I think removing
"Man-" is a reasonable idea. Just to also note that this is not a
problem in labels from other languages. For example, it is not in Greek.
Therefore, it is not a matter of bias on gender in the SIG, but a
problem with the English language/current interpretation of "man" which
we should not ignore.
Posted by Robert Sanderson on 12/4/2019
Dear all,
On behalf of the Linked Art consortium, I would like to propose that the labels for E22 Man-Made Object, E25 Man-Made Feature and E71 Man-Made Thing be changed to drop the unnecessarily gendered “Man-“. In this day and age, I think we should recognize that inclusion and diversity are core features of community acceptance, and that including gender-biased language is alienating.
Thus the proposal is: E22’s label should be changed to Made Object, E25 changed to Made Feature and E71 changed to Made Thing.
The “human” nature of the agent that does the making is explicit in the ontology, in that only humans or groups there-of can be Actors and carry out Productions or Creations, so there is no ambiguity about non-humans making these.
This issue was discussed at length, and has been open in our profile’s tracker for 12 months now. We would greatly prefer that it be solved by changing the labels in the documentation, and thereby in the RDFS, rather than other RDF specific approaches such as minting new terms and using owl:sameAs to assert equality, or rebranding only in the JSON-LD serialization but persisting in other serializations. The change is consistent, reduces the length of the class names, and is an easy substitution. The comprehensibility of the label is still the same. Given the renaming of Collection to Curated Holding, migration of existing data has the same solution - just substitute the labels.
As a second choice, if the above is not acceptable, we propose to instead replace “Man-“ with “Human-“ … only two additional characters, but a bit more of a mouthful.
Many thanks for your engagement with this issue!
Posted by Thanasis Velios on 12/4/2019
I support the change of the English labels to:
E22 Made Object
E25 Made Feature
E71 Made Thing
And I think this can be proposed as an issue to be voted through the SIG
list.
Posted by Pierre Choffé on 12/4/2019
This subject is typical of the politically correct attitude of our times and most people (including me) generally avoid getting involved in such discussions - especially on social media where you would immediately get drowned in a flood of insults - and the result is that we have a feeling of consensus on the matter.
Now, we as a community might have a different point of view, starting with the knowledge we have of the origin of the word "man" (please consult the wikipedia page for a brief introduction). Can we please avoid this kind of discussions and leave it to Twitter and Facebook ?
Et pax in Terra hominibus bonae volontatis... (any woman feeling excluded here ?)
Posted by Florian Kräutli on 12/4/2019
Dear all,
Dear Pierre and all,
I strongly disagree. This is not about the origins of the word but of its usage and meaning in present day. The CRM should reflect (changing) knowledge contexts and we as a community should react to and respect developments in the world, and not decide based on our personal opinions about them.
I think this should be put up as an issue and I would vote in favour of either suggestion: dropping ‘man’ or replacing it with ‘human’.
Posted by Dominic Oldman on 12/4/2019
I strongly agree with Florian.
It is simply right to make these changes.
Posted by George on 12/4/2019
Dear all,
I think there is a distinction to be made in the question of whether the language is in fact biased and whether it is perceived as biased. While I would agree with Pierre that there are arguments to be made that it is not in fact exclusive language in principle (and valid counterarguments to be sure), it is in fact taken by many as being biased and exclusive. This in itself makes it exclusive and this is unnecessary and unwanted.
Since a label in the ontology is just a label, and our intention with the label in this case is to give a heuristic to the ontology user in order to point towards non-naturally generated objects (man made object as we have said to now), I think that dropping 'man' from 'man made', does not impede this functionality.
Removing this part of the label, however, can remove an unintended impression of gender bias. This seems to be a functional gain that is compatible with the spirit of CIDOC CRM (view neutral by nature).
Between 'made' and 'human made', I would lean to the latter. 'Made Object' is already at the limit of understandability in English (it also has some unintended connotations of Mafia language). I think maybe 'human made', while sounding awkward in present day English, may be the direction that everyday language will go anyhow. 'Humankind' sounds very natural and more inclusive than 'mankind' certainly. The adjectival form will also follow.
Another concern is how problematic would the translation be. Checking the translations I could find, I did not find a major problem, but it is something to take into consideration.
A serious technical and cost concern for users of CRM would be that existing data encoded with the old URIs will now be incompatible with this new label. That is a significant trade off.
Finally, there is another class (E24) that includes man made. Added below.
E22 Ανθρωπογενές Αντικείμενο
E24 Ανθρωπογενές Υλικό Πράγμα
E25 Ανθρωπογενές Μόρφωμα
E71 Ανθρωπογενές Δημιούργημα
E22-人造物件 (Man-Made Object)
E24-人造实体物 (Physical Man-Made Thing)
E25-人造外貌表征 (Man-Made Feature)
E71-人造物 （Man-Made Thing）
E71 Künstliches
E22 Künstlicher Gegenstand
E24 Hergestelltes
E25 Hergestelltes Merkmal
I, in any case, think it is probably worth making the change -unless the costs to users in real terms is exorbitant - since the existing label can be perceived to be biased and this is wholly unintended by the community which aims to be both neutral and inclusive.
Posted by Richard Light on 12/4/2019
> A serious technical and cost concern for users of CRM would be that existing data encoded with the old URIs will now be incompatible with this new label. That is a significant trade off.
Ah, but if we had id-only URIs it wouldn't be an issue. Just sayin' ...
Posted by Øyvind on 12/4/2019
Dear all,
I support the change and would also like to point out that this is a local problem with the English language. For instance, in most other Germanic languages the distinction is clear, such as in German: Mann / Mensch or in Scandinavian where we have various versions of mann / menneske.
As for the specific label to be chosen, I leave that for the native English speakers.
Posted by Christian Emil on 12/4/2019
Aas Øyvind points out, the debate is the result of a deficite of The English language. In Swedish for example, the word for 'human' has femine gender.
I have no problem with man-made -> made as long as 'made' is not too wide and include object not made by humans. I checked OED adn it seems ok. But, please check this with somebody with somebody with the right Englsih language expertice. It is not allways so that the natives know their language in this respect.
Posted by Esther Chen on 12/4/2019
Dear All,
I absolutely agree with Florian.
Posted by Franco on 12/4/2019
I (almost) fully agree with Christian-Emil.
But just “Made” could be a misleading label as per se it would include also the result of a deliberate action by my cat: Made Feature = “scratch made on this precious painting by Agatha (the cat) while sharpening her nails”. Instead the scope note indicates it must be the result of human action.
As regards the sexist use of “man":
In Latin “homo” designates any human, the male homo being “vir” versus “mulier”, the female homo: see e.g. “homo sapiens Linn.” and the like.
This use has remained in Latin languages, even if the word “vir” as substantive was sometimes lost: the word derived from homo in modern languages may indicate a human being, regardless of gender, as well as a male of this species: the generic use is a remnant of Latin, not a sexist attribution.
This is the current use in Italian.
I am not sure about Romanian; for French, there is the famous Musée de l’Homme in Paris, which I suspect hosts artefacts concerning both genders. A possible prevalence of Male-Made ones, for the well-known historic reasons, is not why it is called it the “Man Museum".
The Royal Spanish Academy defines “hombre” as "Ser animado racional, varón o mujer” i.e. “Living rational entity, man (varón) or woman (mujer)”. This language kept the Latin distinction even if in the Tex-Mex language “hombre” is usually referred to males only. Interesting to notice that varón does not derive from vir and was originally a derogatory term, this time attributed to males.
In conclusion, this is a matter concerning some Anglo-Saxon allergy caused by the semantic poverty of the language. I would let them go their way and choose whatever they like best, man or human; in the meantime, continue translating it with the gender-neutral term we use in our richer languages.
A label is just a label, so check the implementation cost of the change beforehand: standards are international, not English, so if a bias is perceived by English speakers it is their problem, not mine. Thus out of courtesy I may try to avoid any inconvenience, but I would object paying for the necessary adjustments. On this regard, look at this: https://dilbert.com/strip/2019-04-08.
Franco
PS I did not know the American English use of “Made man” as a Mafia member; here we use the term “initiate” for a person inducted into the Mafia.
Posted by Pierre Choffé
Dear all,
I totally agree with Franco's analysis and so let it be, since we are talking about a deficit of The English language as pointed out by Øyvind. Beware though, this is just the beginning, next step might be renaming anything that has to do with patrimony, and so on and so forth.
Yes Franco, Musée de l'Homme is a good target, according to estimates we should find a compromise with Musée de Tout.e.s (the trendy inclusive Newspeak).
Posted by Achille on 12/4/2019
Dear all,
I fully agree with Franco.
Just a quick remark from the Linguistic world: in order not to run into incorrect analysis of linguistic facts it is always good to keep in mind that the grammatical coding of a language does not reflect the cognitive perception or the cultural and moral judgment of the society/ies that speak(s) this language.
The Germans do not believe that women are things just because “Weib" is a neutral noun rather than a feminine one.
It would therefore be appropriate to keep the linguistic facts (as well as the CIDOC CRM label) sheltered from the cultural facts of our world. But, if the issue is to avoid the opinions of speakers on their own language, why do we not replace English with Latin? This would definitely solve the problem
Posted by Robert Sanderson on 12/4/2019
Hi Franco, Christian, and Pierre,
I agree that this is a modern and solely political issue limited to the English language. English does not have the stability of (said as a native speaker) better languages, nor the Academies that try to keep social pressures from mutating it ad nauseum.
That said, those are the properties of the evolution of the language, and in the current stage of that evolution, it is less and less socially acceptable to use gendered terms when non-gendered would suffice. I expect we can all think of other words in our respective mother tongues that started out completely innocuous and have changed meaning to where the usage has been significantly different.
I agree that there is a technical cost but one that seems less extensive than, for example, deprecating a class or property completely, which happens more often. The trade-off of readability of the terms in RDF compared to the non-linguistic numbers, and the choice of English as a common technical language for that readability, makes this cost unavoidable at times.
For what it’s worth, we also considered “Artificial” but the second sense (insincere or affected) was cause enough to not propose it.
Posted by Jane Sledge
I agree with Robert and Athanasios. Surely the gender of the maker is recorded elsewhere in the CRM and that it does not need to be insinuated as part of E22, E25, and E71? If a distinction is needed between human made and animal made—(a bird’s nest for example), is this now recorded elsewhere in the CRM?
Posted by Franco on 12/4/2019
I don’t want to be invasive of this discussion, just state that also deprecation for the sake of deprecation sometimes seems to me an expensive attitude, although apparently very fashionable in recent times. We theorists often do not take into account the practical implications and the costs related to our vagaries, which we believe to address something fundamental that probably is not so.
As concerns “artificial", what about "Artificial Intelligence”, currently one of the most used buzzwords? Perhaps maliciously we use it, tongue in cheek, thinking to its second sense, “insincere”
Posted by Martin on 12/4/2019
Dear All,
I would like to stay neutral in this issue. Personally, I do not believe that changing language is the way to make sure we respect men and women equally and give them equal chances, and it gives me a taste of distracting from what should be discussed. Therefore I am not happy about it.
I have the impression that even the etymology given in wikipedia https://en.wikipedia.org/wiki/Man_(word) is not complete. As a German speaker, I distinguish between "Mann" (male) and "Mensch" (human), and I suspect that the English "man" is actually a derivative of both, rendering it a homonym. Homonymity would not imply a bias.
In Italian, French, Spanish the Latin term "vir" for adult male actually got lost in favor of derivatives of "homo" (human). Would be interesting to learn if this was actually connected with an increasing male domination or not, or if being "vir" became unimportant.
Man-Made appeared to me a good, established term, and we prefer established terms.
In German, we rendered it as "artificial object".
Asian languages such as Chinese and Japanese do not have default gender at all. Much better.
Anyway, if some people think it makes a difference...
Posted by Christian Emil on 13/4/2019
Dear all,
Martin's reflections are good.
An extra comment: As we all know language use reflect social and power structures. It is my view that one should try to make the language (English) in this case as gender neutral as possible. In many languages gender neutrality is difficult to obtain, perhaps not possible since it is integrated in the inflection morphology (e.g. Russian).
However, there is no reason to make CRM labels gender specific except for mother and father. In English ‘man’ can be used in the meaning ‘humanity’ and unspecified persons (as far as I understand it) which is not very good for gender equality. Maybe one should try to replace ‘man’ by ‘human’. This is a big task, and perhaps not possible in many groups (like the one represented by J R-M). In the group of CRM users it should not be problematic.
The question is: Should we replace ‘man-made’ by ‘human-made’, or by ‘made’. ‘human-made’ is already in use (‘The planet's average surface temperature has risen about 1.62 degrees Fahrenheit (0.9 degrees Celsius) since the late 19th century, a change driven largely by increased carbon dioxide and other human-made emissions into the atmosphere’ NASA) and stress the fact that humans are involved. Rob mentioned that ‘human made’ is quite a mouthful. Well, it will add one syllable and two letters which is not very much.
Posted by Florian Kräutli on 13/4/2019
I would lean slightly in the direction of human-made. I think it helps to emphasise the human activity involved. As Jane mentioned, animals can make things to.
If I remember correctly I already used the term ‘Human-Made Object’ to talk about E22 in this webinar (https://dh-tech.github.io/workshops/2018-10-15-CIDOC-CRMbyPractice/).
Posted by Robert Sanderson on 16/4/2019
Dear all,
It seems like there is general (if not unanimous) agreement around “Human-Made” as a replacement for “Man-Made”?
Can we progress to a vote, or does this need to be discussed in person in June?
Posted by Daria Hookk
Anyway, I propose Hand-Made.
Posted by Melanie Roche on 23/4/2019
Dear all,
I strongly support Robert's suggestion, and thank him for making it.
To answer Martin, I do believe language awareness makes a difference in the real-world situation of men and women. And while it may be a distraction from the work (I totally get that inclusivity is not the point of the model), and appear as a politically correct and fashionable trend, it also reflects on how other communities perceive us. Lately I have heard a lot of discussion on how to expand the CIDOC-CRM community, and especially how to extend it to younger people. Making the model more inclusive would be, in my opinion, a good first step in that direction.
Personally I would rather lean towards "human-made" than simply "made". While "human-made object" is indeed quite a mouthful, "made object" sounds weird and reminds me of "made men" and wiseguys (but maybe I've been watching The Deuce too much). Anyway, either "human-made object" or "made object" sound better than "man-made object".
Posted by martin on 24/4/2019
Dear Melanie,
I totally respect and like your position, but here I have been misunderstood completely:
I meant the discussion about gender neutral language may distract from discussing how to deal in reality with men and women not only as equals, but in deep mutual respect and actually doing it. Using "politically correct" language is not per se respecting, but just silencing.
I was impressed by https://www.amazon.de/Respekt-Heimweh-Menschlichkeit-Renan-Demirkan/dp/3...
For German readers, she explains the difference between tolerance and respect.
See for instance: https://en.wikipedia.org/wiki/Emmy_Noether
"As one of the leading mathematicians of her time, she developed the theories of rings, fields, and algebras. In physics, Noether's theorem explains the connection between symmetry and conservation laws.[4]"
"In 1932 Emmy Noether and Emil Artin received the Ackermann–Teubner Memorial Award for their contributions to mathematics.[70] The prize included a monetary reward of 500 Reichsmarks and was seen as a long-overdue official recognition of her considerable work in the field. Nevertheless, her colleagues expressed frustration at the fact that she was not elected to the Göttingen Gesellschaft der Wissenschaften (academy of sciences) and was never promoted to the position of Ordentlicher Professor[71][72] (full professor).[34]"
Her work is foundational to modern physics. Nevertheless, I have not heard her name in my studies at the university. I heard that her work was later partially attributed to male successors.
The CRM is intended to be factual. Representing provenance of knowledge may be a method to reveal such things. Replacing man-made by human-made is not.
Inclusivity is definitely a goal of the model, and an ethical discussion is not a distraction from "work".
Anyway, I'll send out the Vote. By the way, any SIG member can issue a call for vote.
Posted by Øyvind on 24/4/2019
Dear all,
The discussion of man or human is connected to a larger discussion about changes in natural language.
CRM is meant to have a long life span. Thus, it will see a number of changes in the technologies and artificial languages in which it is expressed, as we discuss regularly, and which we generally adopt to outside the standard itself.
But CRM also makes an anchoring to the human lifeworld through natural languages. The meaning of those languages change over time. We might argue whether the meaning of man in English has changed the last decades or not, but the fact that words change remains. Two examples from my own lifetime:
”Lap” used to be a neutral scholarly and general term for what is now called ”Sami.” Today ”Lap” is more or less exclusively used as a derogatory term.
Until a few decades ago a German term ”Fräulein” referred to an unmarried woman. Today it is never used, at least among the people I meet. I naturally refer to 17 year old female students as ”Frau.” My using ”Fräulein” would at best be seen as a misplaced attempt to be funny or a mistake from an uninformed foreigner.
This means that over some decades the natural language text of CRM will change even if the characters remain the same. Furthermore, we cannot know which words will fare as ”Lap” and ”Fräulein” did. Thus, I suggest that we will, from time to time, come in situations where we have to change the wording of the standard. CRM is, after all, not a novel which should be read in its original language form.
While the case of man vs. human can be attributed to the somewhat unclear term ”political correctness,” others cases will not. They will just be examples of words changing meaning or ceasing to be part of one of the natural languages in which CRM is expressed.
In most cases this will occur in places where it does not matter much, but from time to time it might also happen to labels. This is a cost of using natural language (which we have to) and of the long time span of the CRM (which is one of the points with the whole thing).
Posted by Martin on 24/4/2019
Dear All
The proposal is to replace in all CRM labels "Man-Made" by "Human-Made"
Please vote "YES" if you agree, "NO" if not,
by: Mai 10, 2019
E-VOTES
24/4/2019
Robert Sanderson : YES
25/4/2019
Tony Gill: I vote YES on "human-made." Admittedly it sounds a bit linguistically awkward and cumbersome today, but this change will help the CRM stay on the right side of history going into the future.
Maximilian Schich, Francesco Beretta , Øyvind Eide, Florian Kräutli, Christian-Emil Smith Ore, Richard Light, Esther Chen, Thomas Haensli, Tetsuro KAMURA , Christos Papatheodorou, Annabel Enriquez, Daria Hookk, Maja Žumer : YES
26/4/2019
Athanasios Velios: YES
30/4/2019
George Bruseker: YES
2/5/2019
Mélanie Roche: YES
TOTAL 18 votes: YES to replace in all CRM labels "Man-Made" by "Human-Made"
In the 44th joint meeting of the CIDOC CRM SIG and ISO/TC46/SC4/WG9; 37th FRBR - CIDOC CRM Harmonization meeting, the sig ratified the result of the e-vote regarding the proposal to substitute “Man-” based terms for “Human-“ based ones, instead. 18 members of the sig voted in favor of the proposed change and the decision was unanimous.
All affected classes are to be named accordingly.
The issue is closed.
Paris, June 2019