Issue 488: Modelling an Actor carrying out an action at the Behest of Another

ID: 
488
Starting Date: 
2020-04-14
Working Group: 
3
Status: 
Open
Background: 

Posted by George on 14/4/2020

Here is a minor modelling issue which may or may not find your interest in these times of quarantine. The modelling conundrum is the following:

Sometimes in an activity, the activity is carried out by a named individual but it is carried out on the behest of an organization or someone acts in their capacity as the representative of an organization.

Examples: 

The Conceptual Modelling (E7) is carried out by George Bruseker (E21) as representative of Takin.Solutions (E74)

The Diplomatic Reception (E7) is carried out by Ms. Diplomat (E21) as representative of the Canadian Government

Ie: we want to say that at this time (when E7 occurred) this actor (E21) did the action (p14) but also to qualify this participation not through a role but to say that this person was not acting as an individual but at the time was employed by, working for, acting on the behest of some other entity E74.

One the one hand you could say, just document that the actor was a member of some group through a join and leave event and then you could calculate that they were a member of that group at the time of the event. I think this doesn't work because a) it is obtuse and b) it cannot be inferred that because I am a member of some group at some time that the actions I take in that time span are then me acting on behalf of that group. Obviously, I guess. 

So next potential solution. I think that p14.1 in the role of, won't cut it, because that would only point to a role 'diplomat' 'conceptual modeller' whatever. This does not create the relation to the instance of E39 actor which the E21 acts on behalf of/under the auspices of.

You can't just say that the E21 p107i is current or former member of E74 because a) acting on behalf of someone else doesn't necessarily imply membership in a group together and b) this will not say that the person _at that time_ was acting on behalf of / in relation to the other Actor anyway (see above).

A classic solution might be to create a one person E74 group called 'representative of x organization on this night' and then put the person in that group and then have the group carry out the action. While logically it sounds like a solution (and doesn't call for new additions to the model) but it would be counterintuitive to a user, creating entities that the user wouldn't imagine to think of or use.

Another option would be to do event partioning and then say that the person participated in a sub activity in which they were 'representing' x. I also think this creates a lot. of complication and is not self explanatory as a modelling solution (half the time you should look for actors carrying out the activity under p14 and half the time under a sub event of E7 with a special type). 

So I don't find any of my imagined solutions very satisfactory. What do other people think? Does anyone have a solution that I haven't thought of with existing CRM mechanics? If there isn't a pre-existing solution, do you ideas on how to cover this scenario? 

I encounter it relatively frequently.

One solution I could imagine would be a new .1 type property off the PC14 class that would be something like 'as representative of'. I am not wedded to such a solution, but I suggest it because I think it might link to a more general issue that it is difficult to express 'manner' in a grammatical sense with CRM and somehow the .1 properties aid with this important kind of construct. 

Anyhow just food for thought.

Posted by Pierre Choffe on 14/4/2020

Hi George, hi all,

What about a "Representation Activity", subclass of E7 Activity, that would be "carried out" by the Actor "representing" a Group "in the frame of" another activity ?

Let's say George has 2 activities A1 and A2. Only A2 is carried out as some sort of representation activity.

(George) "carried out" (A1) that resulted in (Something).

(George) "carried out" (A2) that resulted in (Something Else).

(George) "carried out" (Representation Activity) "in the frame of" (A2)

(Representation Activity) "was conducted on behest of" (Group)

Does this capture the semantic of George carrying out an activity as representative of Takin.Solutions ? This would be a light solution which would just require to create a specific event and two associated properties.

All the very best to you and all at CRM-SIG,

Posted by George on 15/4/2020

I agree your proposal sounds like a viable solution to my use case. The idea is not exactly like the sub-activity for breaking down role (which DOREMUS did so well and which Linked.Art adopts as a principle) but, rather that there is a new kind of activity which is 'representing' motivated by these two new properties. So normally I just use p14 (no change so people = happy) and, just in case the activity was motivated by the fact that I am representing somebody (which indeed can be thought of as a separate activity), then I also instantiate a new event and link it as you mention to the actor through p14, to the main event through 'in the frame of' and link the representing event to the group represented by the new property 'on the behest of'. Then perhaps 'in the frame of' could be a specialisation of p17 motivated by and 'was conducted on the behest of' ... don't find a super property for that. It seems like a plausible solution, I guess people wouldn't be very happy for it to appear in CRMbase, but perhaps in CRMSoc. 

I think the alternative of a .1 style property added to p14 is something to be considered as well, because it seems like ontologically it might be more coherent with people's conceptualisation of the situation. I mean to say that it seems like the .1 for in the role of property answers the kind of question which is 'in what way' or 'under what capacity' did so and so do x. Oh, she was responding AS a doctor, he was doing that AS an actor. So I find an analogy here with that kind of construction. George did X AS representative of Y; George did X as actor of kind Q. 

Anyhow they both seem viable solutions, but both require additions to the model. Can anybody see an intuitive way to model this without additions? 

 

Posted by Robert Sanderson on 15/4/2020

Perhaps unsurprisingly, I think the event partitioning pattern is not so unreasonable…

<takin-activities> a Activity ;

  label “all of the activities of Takin Solutions” ;

  carried_out_by <takin> .

<modeling> a Activity ;

  label “modeling activity carried out by George, as a member of Takin” ;

  carried_out_by <george> ;

  part_of <takin-activities> .

<george> a Person ;

  label “George” ;

  member_of <takin> .

<takin> a Group ;

label “Takin Solutions” .

I find this appealing as by acting as a representative of a group, the group also somehow engages with the activity through the partitioning and different carried_out_by references.

Otherwise, what about motivated_by?  The motivation for performing the work is really the *ongoing* membership in Takin, rather than the mere existence of it, but … we know where that leads (cough cough) … so …

<modeling> a Activity ;

  label “modeling activity carried out by George, as a member of Takin> ;

  carried_out_by <george> ;

  motivated_by  <takin> .

It could be motivated by the joining of George to Takin, but that seems less accurate.

 

Posted by Martin on 15/4/2020

Yes, I support this interpretation. There is a question of monotonicity of things described at the Group level with things described individually. This is more important than differentiation between the Group and the individual acting on behalf of it. Can be more detailed wrt the part of link.

 

Posted by George on 16/4/2020

Here's why I'm not sure about these two solutions:

> Perhaps unsurprisingly, I think the event partitioning pattern is not so unreasonable…

<takin-activities> a Activity ;

  label “all of the activities of Takin Solutions” ;

  carried_out_by <takin> .

 

<modeling> a Activity ;

  label “modeling activity carried out by George, as a member of Takin” ;

  carried_out_by <george> ;

  part_of <takin-activities> .

<george> a Person ;

  label “George” ;

  member_of <takin> .

<takin> a Group ;

label “Takin Solutions” .

>I find this appealing as by acting as a representative of a group, the group also somehow engages with the activity through the partitioning and different carried_out_by references.

In the use case that I have at hand, the cataloguer has very little knowledge of what is going on with this person. They know that the person was acting on behalf of this organization at this time, but they may know nothing about the person or the organization really beyond that they have names and these need to be typed into a database. They just have strings that they need to type in and hopefully researchers later can figure something more out. They are, furthermore, not documenting the activities of a company but an object where incidentally this person, representing some organization, is related to it via some sort of activity. So what we have is for example something like:

Bob, representing mega art corp x, carried out a condition survey on object y. This is the factual state of affairs that can be read from some sort of analogue form and must be transported faithfully into the database and then properly represented semantically. 

This can (partially) easily be represented with standard modelling like this:

E22 HMO ->p34i was assessed by -> E14 Condition Assessment

E14 Condition Assessment -.P14 carried out by -> E21 Person

the 'representing mega art corp x' part is the problem. 

In the above proposal they should continue this path with

E14 Condition Assessment->p9i forms part of ->E7 Activities 'The whole activities of mega art corp x'

E21 Person -> p107i was former or current member of E74 Group Mega Corporation X

I think that the E7 Activities node for the 'whole activities of mega art corporation x' which we would make the cataloguer insert here both would be outside the scope of their knowledge and a forever useless node because who would think to find it and use it for anything? It could only become helpful if there was also full documentation around the activities around the corporation which one could reference. So it would be semantically correct but irritating to the data entry user and ultimately unhelpful in helping someone find something or draw a conclusion (that node would never be reused, every time somebody entered more information they would make another 'whole activity of company x' node.

> Otherwise, what about motivated_by?  The motivation for performing the work is really the *ongoing* membership in Takin, rather than > the mere existence of it, but … we know where that leads (cough cough) … so …

<modeling> a Activity ;

  label “modeling activity carried out by George, as a member of Takin> ;

  carried_out_by <george> ;

  motivated_by  <takin> .

> It could be motivated by the joining of George to Takin, but that seems less accurate.

I'm also not convinced by this solution for two reasons. 

a) motivated by is a vague high level property. I use it a lot, but the notion of acting on the behest or as representative is not that vague, it's a specific kind of function. This is okay though, I guess one could create a sub property of motivated by (whether the SIG wanted such a property or not), so that accurate modelling occurred BUT... and this is a big caveat:

b) the motivation is associated to the whole event. So we say that this event of conservation undertaken by star conservator 'bob' on behalf of 'mega art conservation corp x' is motivated by 'mega art corporation x.' However, this would often not be true to the situation. Here are two cases:

 Situation 1: 

E14 Condition Assessment ->p14 was carried out by .E21 Person "Bob the superstar conservator"

E14 Condition Assessment ->p14 was carried out by .E21 Person "John the local conservator"

E14 Condition Assessment ->p17 motivated by -> "Mega art corporation x"

What's the problem here? Bob is motivated by 'Mega art corporation x' because he wants his paycheque and he wants it from them. John isn't.

Situation 2:

But also there is the following problem. Is the event actually motivated by art corporation x? 

What if the condition assessment has been commissioned because there was a fire? Isn't it motivated by the E5 Event of the fire that caused the damage to the art work? Moreover, mega corporation x is actually working on the case because museum Q hired them. So isn't museums Q the motivation? Sure we can say, 'they are all motivations' and be happy to have integrated data, but here perhaps we have over integrated and there are distinctions that are important to know which are papered over because we did not want to spin up a more accurate property to describe the actual situation?

Again, I imagine it's not a CRMbase issue, but for transactions, I have often run into this 'acting for someone else' and still don't see how it would be well described without new classes and properties (haven't found what I'm looking for). 

Posted by Jim Salmons on 16/4/2020

I have been a SIG member for over five years, silent other than my introductory comments upon requesting membership, content to listen and learn from this list while watching the evolution of the CRM. As a post-cancer #PayItForward indie #CitizenScientist, I leave the wrestling with domain-specific modeling issues to you professionals. But when this issue of agency and role-based activities by persons, etc. was raised, I felt an opportunity to contribute a comment based on deep, personal experience with this modeling challenge from my "former life" as an Executive Consultant in the Object Technology Practice of IBM Global Services in the mid-1990s.

So my comments here are not made as a museum informatics professional but rather as a software designer and developer who has wrestled with this issue of role-based agency while trying to "Make it so" in software, as Star Trek's Capt. Picard would say.

Our engagement-based "skunkworks" was inspired by David Gelernter's book, "Mirror Worlds" (https://is.gd/mirrorworlds_at_OxfordUniPress). We explored Gelernter's ideas by creating an "Executable Business Model" framework written in IBM's Distributed Smalltalk. This framework leveraged a role-based metamodel design pattern. After my 2012-13 cancer battle, I was again inspired as I "reinvented" myself as an indie #CitizenScientist doing applied research at the intersection of #DigitalHumanities and #AI/#MachineLearning.

The first focus during my reinvention was on the emerging technology of Open Source graph databases. In graph database technology I saw the promise of non-implementation-specific workflow specifications that could be encoded in a metamodel subgraph of the ground-truth metadata of a digital collection. With these workflow specifications included in the metamodel subgraph partition of a serial publication of, for example, historic magazine and newspaper collections, we could encode not only the collection's complex document structures and content depiction models -- based on an "ontological stack" of #cidocCRM, FRBRoo, and PRESSoo -- we could provide "soft specification" of role-based workflows for both human and machine (micro)services allowed and supported on the digital collection. This ground-truth storage specification is called #MAGAZINEgts and is the subject of my #DATeCH2017 and #DATeCH2019 poster/papers. (PDFs available via my ResearchGate profile: https://www.researchgate.net/profile/Jim_Salmons)

I included a high-level UML Class Diagram in our #DATeCH2017 poster/paper to briefly explain this design pattern, and have uploaded a PNG image of this class diagram to my ResearchGate.net profile here: https://is.gd/cidocCRM_role_agency_modeling. By modeling Person as Actor of Roles, the Actor intermediary class objectifies the expression of agency. Usually, a Person is the Actor of Roles on their own behalf. By providing the explicit Actor, the Person can fulfill the "acting" of a Role by delegation to an Agent, either another Person or a software 'bot'. This Class Diagram shows how a number of #cidocCRM Entities can be laid over a role-based software-executable activity metamodel.

I don't contribute this comment as an alternative to the deep discussion going on here related to this issue from professional CIDOC CRM SIG members' perspectives. I only contribute this here to provide context and a software designer/developer's experience report that I believe is relevant.

BTW, if anyone is interested in my applied research, please do not hesitate to contact me here or via private message. I could use some actual #cidocCRM expertise within my #PLN (Personal Learning Network). I am currently mentored by, among others, Andreas Maier of the FAU Germany Pattern Recognition Lab. Possible collaborations could be expressed in what will become a #DATeCH2021 paper.

From the Cloistered Environs of the #StayAtHome World from the Far Side of The Pond,

 

 

Posted by Martin on 16/4/2020

Dear George, All

Here some more analytical thoughts:

On 4/14/2020 6:47 PM, George Bruseker wrote:

Dear all,

 

Here is a minor modelling issue which may or may not find your interest in these times of quarantine. The modelling conundrum is the following:

 

Sometimes in an activity, the activity is carried out by a named individual but it is carried out on the behest of an organization or someone acts in their capacity as the representative of an organization.

The first thing we should distinguish is if this representation is a permanent or a temporary one.

The second, if the activity is regarded as an activity of the Group in its entirety, or as an activity within the activities of the Group.

For instance, at FORTH, we as researchers are not entitled to speak in the name of FORTH, but the products of our work are propriety of FORTH.

See also the concept of "procura" https://en.wikipedia.org/wiki/Procuration

Generally, if some consequences cannot be inferred from the atomic description, we may need "redundant" modelling, representing respective facts in addition to the causal relation.

 

Examples: 

 

The Conceptual Modelling (E7) is carried out by George Bruseker (E21) as representative of Takin.Solutions (E74)

 

The Diplomatic Reception (E7) is carried out by Ms. Diplomat (E21) as representative of the Canadian Government

 

Ie: we want to say that at this time (when E7 occurred) this actor (E21) did the action (p14) but also to qualify this participation not through a role but to say that this person was not acting as an individual but at the time was employed by, working for, acting on the behest of some other entity E74.

 

One the one hand you could say, just document that the actor was a member of some group through a join and leave event and then you could calculate that they were a member of that group at the time of the event. I think this doesn't work because a) it is obtuse and b) it cannot be inferred that because I am a member of some group at some time that the actions I take in that time span are then me acting on behalf of that group. Obviously, I guess.

Yes, indeed.

So next potential solution. I think that p14.1 in the role of, won't cut it, because that would only point to a role 'diplomat' 'conceptual modeller' whatever. This does not create the relation to the instance of E39 actor which the E21 acts on behalf of/under the auspices of.

We can use two P14 carried out by links, in which the activity is that of the Group P14.1 in the role of : "employer", and P14.1 in the role of: "implementer".

This describes well the incidental connection between employer and employee in actions on behest of the employer.

Note, that all such proposals should be documented, and a good practice of role types be collected!

Alternatively, and compatible with that, would be a description of an overarching activity of the Group, such as "Conceptual Modelling Research" at FORTH ICS CCI, and my individual works as "forms part of" the larger one.

 

You can't just say that the E21 p107i is current or former member of E74 because a) acting on behalf of someone else doesn't necessarily imply membership in a group together and b) this will not say that the person _at that time_ was acting on behalf of / in relation to the other Actor anyway (see above).

Sure.

 

A classic solution might be to create a one person E74 group called 'representative of x organization on this night' and then put the person in that group and then have the group carry out the action. While logically it sounds like a solution (and doesn't call for new additions to the model) but it would be counterintuitive to a user, creating entities that the user wouldn't imagine to think of or use.

I would reject such a solution, because there is no substance of identity to such a one-person Group. This should be restricted to offices and procurators.

In these cases, the relationship is clear.

 

Another option would be to do event partioning and then say that the person participated in a sub activity in which they were 'representing' x. I also think this creates a lot. of complication and is not self explanatory as a modelling solution (half the time you should look for actors carrying out the activity under p14 and half the time under a sub event of E7 with a special type).

That is actually not a problem. See above. I would not use a "representing" x, but "forms part of". I think the concept of an overarching activity is very clear. The "part of" implies the "on behest of". .

We always need to complement search by deductions, inferences between the general and the specific. Therefore "half the time you should look for actors carrying out the activity under p14 and half the time under a sub event of E7 with a special type" is not an argument at all. The closure of all reasonable paths is what people must look at. A good practice guide should enumerate such solutions.

The solution is monotonic in two directions, which is very important:

A) start with the individual activity, not knowing on who's behalf. Adding later a forms part of.

B) start with the Group activity, not knowing the implementer. Adding later the subactivity.

The same holds for the two roles solution above.

 

So I don't find any of my imagined solutions very satisfactory. What do other people think? Does anyone have a solution that I haven't thought of with existing CRM mechanics? If there isn't a pre-existing solution, do you ideas on how to cover this scenario? 

 

I encounter it relatively frequently.

 

One solution I could imagine would be a new .1 type property off the PC14 class that would be something like 'as representative of'. I am not wedded to such a solution, but I suggest it because I think it might link to a more general issue that it is difficult to express 'manner' in a grammatical sense with CRM and somehow the .1 properties aid with this important kind of construct.

I do not like it, because it sounds linguistically nice, but leaves the form of representation under-determined. It cannot distinguish between representing an organisation, such as a procuration, and carrying out a task as employee.

I think we should introduce some specializations of "was motivated by" for commissioning, which has a clear sense. Could be for CRMsoc.

Posted by George on 17/4/2020

Dear Martin,

Thank you for your feedback. Here are some follow up reflections.

So next potential solution. I think that p14.1 in the role of, won't cut it, because that would only point to a role 'diplomat' 'conceptual modeller' whatever. This does not create the relation to the instance of E39 actor which the E21 acts on behalf of/under the auspices of.

We can use two P14 carried out by links, in which the activity is that of the Group P14.1 in the role of : "employer", and P14.1 in the role of: "implementer".

This describes well the incidental connection between employer and employee in actions on behest of the employer.

Note, that all such proposals should be documented, and a good practice of role types be collected!

I don't think this solution fully works because if one is working for one organisation that is doing a job for another one, then one gets

E7 Activity "Lawn Mowing"

pc01i is domain of PC14 Carried Out by pc02 carried out by E21 "George Bruseker"

pc01i is domain of PC14 Carried Out by p14.1 in the role of E55 "implementer"

pc01i is domain of PC14 Carried Out by pc02 carried out by E74 "Conceptual Lawn Mowers R Us"

pc01i is domain of PC14 Carried Out by p14.1 in the role of E55 "Employer"

 

We already don't have a direct connection between the E74 and the E21. I have to mentally infer it. This especially comes out if we add the next actor, the one who is actually paying:

pc01i is domain of PC14 Carried Out by pc02 carried out by E21 "Lee M. Citizen"

pc01i is domain of PC14 Carried Out by p14.1 in the role of E55 "Employer"

Now I have no machine decidable way of deciding who is employing who, but I suppose it would be a reasonable documentation scenario to put just exactly this.

Alternatively, and compatible with that, would be a description of an overarching activity of the Group, such as "Conceptual Modelling Research" at FORTH ICS CCI, and my individual works as "forms part of" the larger one.

 This sounds like the suggestion of Rob which is great if that kind of analytic documentation is being carried out, but in many scenarios it will not be. The individual doing the documentation can only note the facts that are before them. A represented B. They could, in principle, also create event records for the total activity of an organization, but in principle they are probably even loathe to document separately the organization itself qua entity. So while the model may be sound semantically, it takes a distance from actual documentation practice.

Another option would be to do event partioning and then say that the person participated in a sub activity in which they were 'representing' x. I also think this creates a lot. of complication and is not self explanatory as a modelling solution (half the time you should look for actors carrying out the activity under p14 and half the time under a sub event of E7 with a special type).

That is actually not a problem. See above. I would not use a "representing" x, but "forms part of". I think the concept of an overarching activity is very clear. The "part of" implies the "on behest of". .

We always need to complement search by deductions, inferences between the general and the specific. Therefore "half the time you should look for actors carrying out the activity under p14 and half the time under a sub event of E7 with a special type" is not an argument at all. The closure of all reasonable paths is what people must look at. A good practice guide should enumerate such solutions.

I think frustrated software designers and non semantic savvy end users (almost everyone) may see this issue less expansively. 

The solution is monotonic in two directions, which is very important:

A) start with the individual activity, not knowing on who's behalf. Adding later a forms part of.

B) start with the Group activity, not knowing the implementer. Adding later the subactivity.

The same holds for the two roles solution above.

In the real scenario I describe, the problem is not that the documentarist does not know on whose behest the person is working, it is that they do not and are not going to create separate records of the entire activity of an organization separate from that organisation in order to make the CIDOC CRM modelling work. So in order to use the part of type modelling proposal they need to create a node for the 'whole activity of organization x' which is in fact a useless node because it will be a blank hidden node that will get recreated many times to satisfy the model, but will in the knowledge base just create endless duplicate 'whole activity of organization x' nodes which not only don't help retrieval but hinder it. 

 My general problem with the portioning solution, however, has two parts.

a) if we endorse both the p14.1 in the role of and the partitioning by E7 Activity for this very basic human and documentation phenomenon, then in CIDOC CRM it is as if we have created irregular verbs. So it is like having created Esperanto but then immediately adding in irregular verbs to spice it up. To me this is not ideal. A sentence of the form 'Bob played the role of Y in Activity X' should ideally have one translation in CRM (or a long cut and a shortcut which are logically equivalent). Perhaps it is too purist.

b) the partitioning solution may get called on to do too much. So I think of the DOREMUS example, which is great. There, because the actions of each individual in the performance have so many variables, a partitioning is performed so that each player in the orchestra performs their own individual activity within the whole.  Beautiful. The principle of partitioning is also clear. Ah, but what now, if in my dystopian present, some of the players in the band are commissioned or paid for, or stand for some other organization. The 'Pepsi sponsored violinist'. Now I must also do partitioning in this overall orchestral act but whereas before I had one principle of partitioning (role as musician in the musical whole of the performance) now I have two (role as brand sponsor in advertising campaign). So Sally the violinist whom plays the first violin in one act must in a different act (or in a sub act of this act, or maybe the sponsorship is the overall act of her sub act and her performance the sub activity of the sponsored activity) perform the being sponsored relationship.

So I don't find any of my imagined solutions very satisfactory. What do other people think? Does anyone have a solution that I haven't thought of with existing CRM mechanics? If there isn't a pre-existing solution, do you ideas on how to cover this scenario? 

I encounter it relatively frequently.

One solution I could imagine would be a new .1 type property off the PC14 class that would be something like 'as representative of'. I am not wedded to such a solution, but I suggest it because I think it might link to a more general issue that it is difficult to express 'manner' in a grammatical sense with CRM and somehow the .1 properties aid with this important kind of construct.

I do not like it, because it sounds linguistically nice, but leaves the form of representation under-determined. It cannot distinguish between representing an organisation, such as a procuration, and carrying out a task as employee.

I think a common scenario (the typical knowledge state of the documentarist in the culture heritage institution) would be of having this level of underdetermined knowledge (which is yet knowledge!) and only very particular research being able to recover more details. So this could be the first pass of knowledge and then if researchers came in they could document the activities of the group that is being represented and then eventually contracts signed that lead to the particular motivated activities, but in the meantime this basic level of knowledge captured. 

Again I'm not wedded to the solution, but I do see problems with the above formulations, at least for the scenarios I have at hand.

Posted by George on 17/4/2020

Thanks for your thoughts on the matter. So, do I read your modelling correctly that you make of a role class that is a subtype of E74 group, or do I not read that right? If so it reminds me of the solution come up with by a group that made what they called CRMbio (I can't find the reference right now, but it's out there!). To handle roles, they simply argued that activities were not undertaken by people or groups but by roles (I think I represent their thinking more or less correctly like that.... if they are on the list and I'm wrong, please correct me). What I didn't find convincing about this is that CRM attempts to create a tool that is correlative to real world entities and how we speak about them actually behaving. To say that activities are carried out by roles kind of works theoretically and has some technical advantages, but is it how we actually view the world? Or do we think that people and groups do the acting and roles are a (perhaps ultimately conditioning) function of their action? 

 

Posted by Martin on 22/4/2020

This may find your interest:

F. Steimann. On the representation of roles in object-oriented and conceptual modelling.Data& Knowl-edge Engineering35(1): 83–106, 2000.

This is a back ground paper of the current CRMbase approach.

I found these, but have not yet read in detail:

http://ceur-ws.org/Vol-2205/paper25_ontocom4.pdf

and particularly

https://books.google.gr/books?id=n3cRDAAAQBAJ&pg=PA74&lpg=PA74&dq=Ontology+of+delegation+of+action&source=bl&ots=_ozargCAze&sig=ACfU3U2aV028Cvm0Ts_64ieVHVcsmfQ51w&hl=el&sa=X&ved=2ahUKEwjS2Lbf2_zoAhVFlFwKHchzAoIQ6AEwC3oECAgQAQ#v=onepage&q=Ontology%20of%20delegation%20of%20action&f=false

Social Ontology: Collective Intentionality and Group Agents

Εξώφυλλο

Raimo Tuomela

Oxford University Press, 23 Αυγ 2013 - 320 σελίδες

 

Social ontology, in its broadest sense, is the study of the nature of social reality, including collective intentions and agency. The starting point of Tuomela's account of collective intentionality is the distinction between thinking and acting as a private person ("I-mode") versus as a "we-thinking" group member ("we-mode"). The we-mode approach is based on social groups consisting of persons, which may range from simple task groups consisting of a few persons to corporations and even to political states. Tuomela extends the we-mode notion to cover groups controlled by external authority. Thus, for instance, cooperation and attitude formation are studied in cases where the participants are governed "from above" as in many corporations. The volume goes on to present a systematic philosophical theory related to the collectivism-versus-individualism debate in the social sciences. A weak version of collectivism (the "we-mode" approach) depends on group-based collective intentionality. We-mode collective intentionality is not individualistically reducible and is needed to complement individualistic accounts in social scientific theorizing. The we-mode approach is used in the book to account for collective intention and action, cooperation, group attitudes, and social practices and institutions, as well as group solidarity. Tuomela establishes the first complete theory of group reasons (in the sense of members' reasons for participation in group activities). The book argues in terms of game-theoretical group-reasoning

Old Proposal: 

Posted by Francesco Beretta on 23/4/2020

Dear George, Martin, Rob, all

Thank you for this very interesting and relevant discussion which definitely belongs to CRMsoc. I'd kindly ask those who can, to create an issue in the CRMsoc documentation, with these emails, in order to make this discussion more accessible. Also, in my opinion, this rich discussion shows the limits of a mailing list: it would be very useful to split and regroup the different sub-questions and answers within different threads (e.g in a forum) and it takes a lot of time to read and reorder all the points of view — in the own mind or on 'paper'. In the end, only a few people will have the time and the motivation of doing this, while the interest of the issue would deserve discussion by a wider community.

This said, I think three different levels appear and are partly mixed up: a phenomenal, an epistemological and a technical. And this makes the issue even more difficult to solve, at least to the extent that these different levels are not differentiated.

On the phenomenal level the question is: what is the modelled phenomenon ? A personal, time-related quality or skill of the person in charge of the activity ? or the fact that he/she acts as representative of an institution, as a more general activity ? or with a specific mission in this case ? or because he/she is employed by an organization and carries out that activity within that framework ? It seems difficult to have a unique way of modelling all these different possible aspects of reality.

Also, the perception of them depends on the point of view of the observer, as social sciences teach us. Even in natural science, objectivity is a matter of convention and the model of reality is only one of the possible representations of it, not yet falsified. This is even more true for social phenomena, even if one limits oneself to the level of pure information. Choosing between phases, time-limited qualities of entities or events to model these social facts is therefore as much the result of epistemological choices as it is the result of the comtemplation of the phenomenal reality as such. Definitely an issue for CRMsoc where the epistemological approach should be wider then the one in CMRbase. Assuming that the modelled domain is the one of social states of affairs.

 

And finally there is the techical issue. We try to model this complex reality, and all these different perspectives, with simple, limited constructs like (RDFS) classes and properties, then —given the richness of the phenomena— we are obliged to introduce additional constructs, such as properties of properties (14.1 etc.), property classes (PC) or by splitting events in sub-events through partitioning, which are not really specified in the standard, or at least not in a very visible way for the community.

 

During the 12 years of the symogih.org experience we had long discussions on this issue (without beeing able to really answer it) : knowing that a person is involved in an event and has thus a role in it, are the aforementioned phenomena characteristic of the person, of the role, or of both in the context of that event ? the answer depends on the modelled phenomenon and on the point of view of the data producer.

 

Technically speaking one could express this in (at least) two ways:

 

1.

 

<modeling> a Activity ;

    label “modeling activity carried out by George, as a member of Takin> ;

    carried_out_by _[actor-with-contextual-quality]*.

 

_[actor-with-contextual-quality] has_actor <george> ;

    has_quality <time related skill>**.

    [or]

    has_motivation <specific mission for this activity>**.

    [or]

    has_general_activity <more general activity>**.

 

 

* blank node

** the corresponding temporal entities, with own properties or (if shortcuts and simplifications) the corresponding types

 

 

2.

 

<modeling> a Activity ;

    label “modeling activity carried out by George, as a member of Takin> ;

    carried_out_by* <george>.

 

carried_out_by*  with_the_quality** <time related skill>.

[or]

carried_out_by* with_the_motivation** <specific mission for this activity>.

[or]

carried_out_by* in_the_contex_of_general_activity** <more general activity>.

 

 

* as PC or reified property (I do not use here the usual statement construct for reified properties to keep it readable)

** as  property of property

Solution 1. focuses on the quality or mission of the actor but raises the question of the identity of the blank node, as stated in the previous discussion on this list. A blank node has not a specific identity but how are then defined the related properties ?

This approch expresses in a suitable manner the social quality inherent to the actor, whether perceived or factual, occurring mainly during the activity. It is therefore nearer to reality or, at least, our discourse about reality.

Solution 2. emphasizes the importance of the actor's role in the context of the action, qualifies and clarifies it. It adopts an existing construct (statement reification) but calls for a clearer definition in CRM and its model family of the meaning of 'properties of properties' and their use. And also: in fact the quality or mission does not belong to the role, but to the actor, so this kind of modelling is somewhat artificial.

Both solutions seem to work technically but reveal the difficulty of expressing a complex reality and specific points of view with simple constructs.

 

Posted by Martin on 23/4/2020

Dear Francesco,

I support very much your arguments. I currently see at least 5 distinct cases, I'll summarize the next days.

Just a quick remark: Please do not use the label "general" in this sense: "has_general_activity <more general activity>**.", because we have conflicting interpretations, and this is an open issue.

Use "has wider activity"  or "extended activity" or so.

Chrysoula will assign an ISSUE number.