Agency, Capabilty, Role, realisation
Thomas Beale
Barry Smith
I've been having a long look at how to coherently integrate the very common business / info modelling concept of 'party', which I can best describe as an 'entity with agency', meaning an entity that can in and of itself make decisions, plan and actively participate in relationships.In business modelling, kinds of Parties are typically Person, Automaton, Organisation, and Group - any autonomous entity that acts in a unified way, either due to being a physical individual (Person, Automaton) or having internal mechanisms that achieve unified action, as companies etc do.Business entity models also tend to treat parties-with-capabilities, e.g. professions such as nurse, accountant, rugby player etc as Parties, such that a Person Jane Smith who has capability of General Practitioner may enter into a relationship with Nuffield Healthcare Trust, to practice as a GP.However, Parties may just be plain old Persons and Organisations - since such entities have civil / legal standing, and may enter into relationships, or perform activities without any requirement for specific capabilities, e.g. Dan Jones hires Pete Smith to mow his lawn.The question is how to consider these notions in BFO terms. BFO doesn't directly address the notion of agency, and tends to treat ways of acting that pre-suppose agency, such as being a nurse as Roles. Being a legal Person would presumably be similar, i.e. distinct from being a biological organism (bfo::object).In BFO, Role is a realisable entity, instances of which are realised in processes of correlated types.
However in reality, we routinely designate someone a 'nurse' if he/she has the appropriate RN certification for the jurisdiction. The same is true for every profession: we can say Joe is a welder, Mary is an artist and Amy is an auditor - without any of them being employed as such, or 'realising' the supposed role of nurse, welder and so on.
Indeed, the whole basis for the labour market is that individuals advertise themselves as nurse, welder, etc etc (with appropriate qualifications) so that they can obtain employment in which they can act as such.However, if the Role of being a nurse only obtains at instance level when there is a) an established employment, and b) a concrete work situation, then it appears difficult to represent a person-as-nurse.
One problem here is that the term 'role' is mis-used by nearly everyone to mean at least three things:A) the capability to work in the role of a nurse, GP etc
B) employed in the role of a nurse at xyz health practiceC) actually working as a nurse today at 12:15 pm with patient 123 etc
We might think of A (nurse capability) as the proposed BFO capability category (between disposition and function), but this is still a realisable entity requiring a situation (process) for realisation. It's no problem at the level of universals (i.e. 'nurse' as a type) but it wouldn't work at instance level, where we want to point to a Person Jane Smith, RN - not currently working as a nurse.The employment form of 'role' (B) is (I think) bfo::relational quality - representing the employment relationship. The realised form C) is presumably also bfo::role, with participation in a process (the work situation).Somewhat confusingly the Capabilities paper (Merrell et al) treats Capability as a Disposition, but many capabilities of real world interest are gained by education, and could be lost by lack of practice, forgetting etc.
This is may be a side issue, but I can't help thinking that capability cannot be a disposition, for the kinds of capabilities that are acquired (by education, training, practice) and can be lost. And yet, in ordinary usage, some dispositions are clearly 'capabilities'.I would suggest therefore that a new BFO class is needed whose meaning is 'acquired skill' or perhaps 'acquired ability' would be better. Both this class and Disposition would have a new common parent i.e. between them and realisable entity, which could even be named Capability, understood in the sense of 'ability'. All Dispositions would be this kind of Capability, but innate disposition is only one route to capability; acquisition (of any kind) is another.This new Capabilty would still be internally grounded, hence sits correctly in distinction to Role. In summary:Realisable entity+--- Role+--- Capability (understood as ability)+--- Disposition+--- Function+--- Acquired ability (= the 'Capability' of the cited paper)
This still doesn't really address the problem that all Realisable entities are only realised at instance level in a process (= some situation) at least for social / business capabilities, for the simple reason that professions / skills are names for individual abilities, e.g. nursing means that you can do things like insert a catheter, arrange an IV. Indeed, 'nursing' as a verb has only narrow meanings like a) breast-feeding or b) nursing an injury. Even 'welder' as a profession description and 'welding' as a directly realisable activity are not the same - welders also have to know how to dress safely for welding, how to maintain a welding machine, what flux rods to choose etc.I'll just throw in one last spanner: in business systems, individuals represented as digital twins (e.g. in an administration system, master patient index, SAP etc) are only represented in terms of their business selves, not as their material selves. So the digital twin (= BFO individual mapped in data) of Jane Smith, RN, will be a self-standing entity, with no representation of the bearer (which would be a biomedical kind of digital twin of Jane Smith, which will not be in any admin system). Thus, Realisable entities such as Roles and Capabilities (as defined above) are instantiated as self-standing entities, that stand for any assumed bearer. This probably doesn't matter in practical IT system design but it doesn't quite follow the standard (materialist) BFO narrative.
--
You received this message because you are subscribed to the Google Groups "BFO Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bfo-discuss...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bfo-discuss/b07033ab-f6d1-4513-b62f-b5b23d46e493n%40googlegroups.com.
Woland's Cat
On Thu, Jun 23, 2022 at 08:27 Thomas Beale <wolan...@gmail.com> wrote:
However in reality, we routinely designate someone a 'nurse' if he/she has the appropriate RN certification for the jurisdiction. The same is true for every profession: we can say Joe is a welder, Mary is an artist and Amy is an auditor - without any of them being employed as such, or 'realising' the supposed role of nurse, welder and so on.Their role would still be realized in the way described. Many roles (and many capabilities) are never realized.
TB: I have problems with this sentence, because we don't really
know what 'role' means (given 3 interpretations noted below and
you have added a 4th).
Indeed, the whole basis for the labour market is that individuals advertise themselves as nurse, welder, etc etc (with appropriate qualifications) so that they can obtain employment in which they can act as such.
However, if the Role of being a nurse only obtains at instance level when there is a) an established employment, and b) a concrete work situation, then it appears difficult to represent a person-as-nurse.
The documentation makes it clear that a role is externally grounded. To become a nurse you have to have been assigned the role externally typically in some certificate. And that is what matters for getting a job.
TB: I don't really dispute the clarity of the formal definition of 'role' (or anything else) in BFO; however, it's a word that is very slippery in English, and even 'examples' or explanatory text can misuse it.
In the Party
model I am working on (heavily BFO influenced), I got rid of
'role' entirely, it causes too much trouble. I have opted for
'accountability' to mean something like the description of a job
post, and a (party) relationship to express the actual employment
relationship (if and when it occurs).
One problem here is that the term 'role' is mis-used by nearly everyone to mean at least three things:A) the capability to work in the role of a nurse, GP etc
We now have capability for this - I can send you a new draft paper if you likeB) employed in the role of a nurse at xyz health practiceC) actually working as a nurse today at 12:15 pm with patient 123 etcActually there is one more:D) the authority to realize the role of nurse (see oncertificate above) - this is at the heart of the BFO treatment
TB: is this really different from B (employment), since that would normally be based on accountability and therefore authority?
...
Somewhat confusingly the Capabilities paper (Merrell et al) treats Capability as a Disposition, but many capabilities of real world interest are gained by education, and could be lost by lack of practice, forgetting etc.
Many BFO dispositions are just like that
TB: I am mystified by this, since learned capabilities don't have a material basis, unless we want to posit a 2nd (?nth order) argument to do with innate capacity of the the brain, memory etc to learn things? Or has the BFO view of disposition changed in recent times?
Realisable entity+--- Role+--- Capability (understood as ability)+--- Disposition+--- Function
+--- Acquired ability (= the 'Capability' of the cited paper)
Unfortunately there are infinitely many dispositions which are not capabilities
TB: right - in the standard sense of the word, i.e. some positive ability that might be sought in some circumstance.
But understood as a neutral ability (blood coagulating, water
crystallising at 0oC etc) I think there is a discussion to be had.
I agree that 'capability' is is not the correct term in the above,
I was just trying to show that with a wider definition, it could
be retained somewhere in the ontology. I'll just call the more
general term 'capacity' for now (I think that is better than
'ability').
The problem with bfo::disposition is that losing (and gaining) a
disposition means that the bearer is materially changed. That is
what one would expect with physical / material / biological
dispositions. But many 'abilities' (per above) are not achieved by
innate bio-physical make-up, but by acquisition of knowledge. (I
realise I was a bit sloppy in the first post - I was only thinking
of knowledge-based capabilities, not aquired physical changes such
as the sharpened stick example, or physical training giving the
capability of being able to lift heavy weights).
So I would still argue firstly that there is a category of
realisable entity that is internally grounded, but is neutral
with respect to whether materially or informationally so.
Secondly, I would argue that there is a kind of capacity that is
grounded in acquired knowledge, which I will call a 'skill' for
now. And we already have disposition as a materially necessary
kind of capacity. This would give us the following.
I think we could now say that all Dispositions really are Capacities in the sense above.
My problem with the definition of Capability defended in the Merrell et al paper is that it straddles the following disjoint categories:
- a subset of Dispositions, i.e. materially necessary capabilities
- some of which are bio-physically innate, e.g. water freezing at 0degC
- some of which are acquired physical changes, e.g. sharpened sticks etc
- skills, which are not materially necessary
The Capability paper starts with the following statement in section 2.2 The Need for Capability: For our purposes here, a capability is what an entity is able to do in virtue of its material constitution. But the paper later argues the following in section 4.8:
Capabilities are central to education,
training, and even so-called talent acquisition for government
and private organizations. These fields relate to each other
within a system. Organizations both public and private have an
interest in acquiring competent individuals who can further
their goals and agendas. However, to acquire these skills these
individuals need to be educated through school and universities,
and sometimes also trained through institutions such as
vocational colleges and military bootcamps.
Accordingly, capabilities and the grades
achieved in realizing them – expressed in diplomas, warranties,
or baseball statistics tables – are vehicles through which
organizations are able to express and fulfil their needs
So I think if the paper wants to argue that skills are
dispositions in the sense of material necessity, then it would
need to show how this is so. (Indeed it mentions Chemero's
argument as to why capabilities cannot be dispositions, but rebuts
it on a basis that to me implies that disposition is not wholly
materially based. I am more persuaded by CHemero's argument at the
moment! This would also have implications for gradability - see
below).
I would argue that 'capability' is a relative and probably subjective term, i.e. relative to situation, other parties, etc, in which innate abilities are 'of interest'. Therefore I don't think that BFO, being materially based, can be expected to support capability as a clean category.
I also think that gradability applies mostly to knowledge-based
skills, on an individual basis. Gradability of
dispositions that are useful (the kind of capability proposed in
the paper) I think can only be done across a population. Consider
the sharpened stick - you can't really say that it is a grade 5
(out of 10) sharp stick, and if it were 'sharper', it would pierce
things better. The said stick has just one level of sharpness -
the one that it has, and it will pierce things just as well as
that kind of sharp stick will do, no more and no less. That will
not change over time, unless the sharpness of the end is materially
changed somehow. You could only create a sharpness-grading
system by looking at many sticks of differing sharpness.
To me it seems that the Capability proposal could be adjusted to
exclude knowledge-based capabilities from its notion of Capability
- but then it is probably going to need to use a different term
that means something like 'situationally useful disposition' -
maybe go back to 'ability'? I.e.
There is a separate question of whether knowledge-based
capabilities (skills, as I have termed them) could be added to
BFO. If a strictly material basis is required, then probably there
is no place for skills as such. We could even tak about simple
information, such as military intelligence - it's not even a
skill, it's just being in possession of a fact, but it allows one
to kill the enemy before they kill you. This obviously points to a
deeper philosophical question of how far a materially based
ontology can go in addressing knowledge- and information-based
capabilities - both in individuals and in organisations. I am
inclined to think we should be looking elsewhere or even be
thinking about a multi-hierarchy approach. (As an example of why,
see this
part of our draft Party model, which has a hierarchy that
makes sense in terms of agency / business activity, but does not
map cleanly at all to BFO - just consider what an 'organisation'
is materially...)
Finally, I suspect the word 'capability' is better as a useful English word that has no clean BFO counterpart.
Barry Smith
On 27/06/2022 00:45, Barry Smith wrote:
On Thu, Jun 23, 2022 at 08:27 Thomas Beale <wolan...@gmail.com> wrote:
However in reality, we routinely designate someone a 'nurse' if he/she has the appropriate RN certification for the jurisdiction. The same is true for every profession: we can say Joe is a welder, Mary is an artist and Amy is an auditor - without any of them being employed as such, or 'realising' the supposed role of nurse, welder and so on.Their role would still be realized in the way described. Many roles (and many capabilities) are never realized.TB: I have problems with this sentence, because we don't really know what 'role' means (given 3 interpretations noted below and you have added a 4th).
Indeed, the whole basis for the labour market is that individuals advertise themselves as nurse, welder, etc etc (with appropriate qualifications) so that they can obtain employment in which they can act as such.
However, if the Role of being a nurse only obtains at instance level when there is a) an established employment, and b) a concrete work situation, then it appears difficult to represent a person-as-nurse.
The documentation makes it clear that a role is externally grounded. To become a nurse you have to have been assigned the role externally typically in some certificate. And that is what matters for getting a job.TB: I don't really dispute the clarity of the formal definition of 'role' (or anything else) in BFO; however, it's a word that is very slippery in English, and even 'examples' or explanatory text can misuse it.
In the Party model I am working on (heavily BFO influenced), I got rid of 'role' entirely, it causes too much trouble. I have opted for 'accountability' to mean something like the description of a job post, and a (party) relationship to express the actual employment relationship (if and when it occurs).
One problem here is that the term 'role' is mis-used by nearly everyone to mean at least three things:A) the capability to work in the role of a nurse, GP etc
We now have capability for this - I can send you a new draft paper if you likeB) employed in the role of a nurse at xyz health practiceC) actually working as a nurse today at 12:15 pm with patient 123 etcActually there is one more:D) the authority to realize the role of nurse (see oncertificate above) - this is at the heart of the BFO treatmentTB: is this really different from B (employment), since that would normally be based on accountability and therefore authority?
...
Somewhat confusingly the Capabilities paper (Merrell et al) treats Capability as a Disposition, but many capabilities of real world interest are gained by education, and could be lost by lack of practice, forgetting etc.
Many BFO dispositions are just like thatTB: I am mystified by this, since learned capabilities don't have a material basis
, unless we want to posit a 2nd (?nth order) argument to do with innate capacity of the the brain, memory etc to learn things?
Or has the BFO view of disposition changed in recent times?
Realisable entity+--- Role+--- Capability (understood as ability)+--- Disposition+--- Function
+--- Acquired ability (= the 'Capability' of the cited paper)
Unfortunately there are infinitely many dispositions which are not capabilitiesTB: right - in the standard sense of the word, i.e. some positive ability that might be sought in some circumstance.
But understood as a neutral ability (blood coagulating, water crystallising at 0oC etc) I think there is a discussion to be had. I agree that 'capability' is is not the correct term in the above, I was just trying to show that with a wider definition, it could be retained somewhere in the ontology. I'll just call the more general term 'capacity' for now (I think that is better than 'ability').
The problem with bfo::disposition is that losing (and gaining) a disposition means that the bearer is materially changed.
That is what one would expect with physical / material / biological dispositions. But many 'abilities' (per above) are not achieved by innate bio-physical make-up, but by acquisition of knowledge.
(I realise I was a bit sloppy in the first post - I was only thinking of knowledge-based capabilities, not aquired physical changes such as the sharpened stick example, or physical training giving the capability of being able to lift heavy weights).
So I would still argue firstly that there is a category of realisable entity that is internally grounded, but is neutral with respect to whether materially or informationally so.
Secondly, I would argue that there is a kind of capacity that is grounded in acquired knowledge, which I will call a 'skill' for now. And we already have disposition as a materially necessary kind of capacity. This would give us the following.
Realisable entity+--- Role+--- Capacity (= any kind of ability / tendency to act in a certain way in specific circumstances)
+--- Disposition (=materially necessary capacity)
+--- Function
+--- Skill (= acquired capacity based in knowledge / learning)
I think we could now say that all Dispositions really are Capacities in the sense above.
My problem with the definition of Capability defended in the Merrell et al paper is that it straddles the following disjoint categories:
- a subset of Dispositions, i.e. materially necessary capabilities
- some of which are bio-physically innate, e.g. water freezing at 0degC
- some of which are acquired physical changes, e.g. sharpened sticks etc
- skills, which are not materially necessary
The Capability paper starts with the following statement in section 2.2 The Need for Capability: For our purposes here, a capability is what an entity is able to do in virtue of its material constitution. But the paper later argues the following in section 4.8:
Capabilities are central to education, training, and even so-called talent acquisition for government and private organizations. These fields relate to each other within a system. Organizations both public and private have an interest in acquiring competent individuals who can further their goals and agendas. However, to acquire these skills these individuals need to be educated through school and universities, and sometimes also trained through institutions such as vocational colleges and military bootcamps.
Accordingly, capabilities and the grades achieved in realizing them – expressed in diplomas, warranties, or baseball statistics tables – are vehicles through which organizations are able to express and fulfil their needs
So I think if the paper wants to argue that skills are dispositions in the sense of material necessity,
then it would need to show how this is so. (Indeed it mentions Chemero's argument as to why capabilities cannot be dispositions, but rebuts it on a basis that to me implies that disposition is not wholly materially based. I am more persuaded by CHemero's argument at the moment! This would also have implications for gradability - see below).
I would argue that 'capability' is a relative and probably subjective term, i.e. relative to situation, other parties, etc, in which innate abilities are 'of interest'. Therefore I don't think that BFO, being materially based, can be expected to support capability as a clean category.
I also think that gradability applies mostly to knowledge-based skills, on an individual basis. Gradability of dispositions that are useful (the kind of capability proposed in the paper) I think can only be done across a population. Consider the sharpened stick - you can't really say that it is a grade 5 (out of 10) sharp stick, and if it were 'sharper', it would pierce things better. The said stick has just one level of sharpness - the one that it has, and it will pierce things just as well as that kind of sharp stick will do, no more and no less. That will not change over time, unless the sharpness of the end is materially changed somehow. You could only create a sharpness-grading system by looking at many sticks of differing sharpness.
To me it seems that the Capability proposal could be adjusted to exclude knowledge-based capabilities from its notion of Capability - but then it is probably going to need to use a different term that means something like 'situationally useful disposition' - maybe go back to 'ability'? I.e.
Realisable entity+--- Role+--- Capacity (= any kind of ability / tendency to act in a certain way in specific circumstances)
+--- Disposition (=materially necessary capacity)+--- Ability (= a useful disposition)
+--- Function
+--- Skill (= acquired capacity based in knowledge / learning)
Secondly, I think the notion of gradability with respect to dispositional abilities needs to be defended better.
There is a separate question of whether knowledge-based capabilities (skills, as I have termed them) could be added to BFO. If a strictly material basis is required, then probably there is no place for skills as such. We could even tak about simple information, such as military intelligence - it's not even a skill, it's just being in possession of a fact, but it allows one to kill the enemy before they kill you. This obviously points to a deeper philosophical question of how far a materially based ontology can go in addressing knowledge- and information-based capabilities - both in individuals and in organisations. I am inclined to think we should be looking elsewhere or even be thinking about a multi-hierarchy approach. (As an example of why, see this part of our draft Party model, which has a hierarchy that makes sense in terms of agency / business activity, but does not map cleanly at all to BFO - just consider what an 'organisation' is materially...)
Finally, I suspect the word 'capability' is better as a useful English word that has no clean BFO counterpart.
--
You received this message because you are subscribed to the Google Groups "BFO Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bfo-discuss...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bfo-discuss/92ed24ed-328b-7090-2245-b5a4367a15a8%40gmail.com.
Woland's Cat
On Mon, Jun 27, 2022 at 10:45 AM Woland's Cat <wolan...@gmail.com> wrote:
On 27/06/2022 00:45, Barry Smith wrote:
Their role would still be realized in the way described. Many roles (and many capabilities) are never realized.TB: I have problems with this sentence, because we don't really know what 'role' means (given 3 interpretations noted below and you have added a 4th).
A role is an externally grounded realizable entity.It does not have to be realized.That is the only sense of 'role' in BFO.
TB: Just to be clear, I am not criticising BFO here, only saying
that the rest of the world uses 'role' in such undisciplined ways
that it's a poisoned word in a precise ontology, and thus a bit
annoying, because one always has to go through the explanation of
'ok, but role in BFO means xyz etc'.
In the Party model I am working on (heavily BFO influenced), I got rid of 'role' entirely, it causes too much trouble. I have opted for 'accountability' to mean something like the description of a job post, and a (party) relationship to express the actual employment relationship (if and when it occurs).
I think accountability is too narrow -- it suggests that the person in question will be held responsible.
TB: That is indeed the intention, and you may be right on it being too narrow. We are not of course trying to create another BFO, so we'll need to see if it trips us up when we work through more examples...
Nevertheless, re: your comment below 'Accountability (in my
reading) involves institutional mechanisms to bring to account' -
our Accountability is a bit wider than that - it is simply one
Party holding another accountable to do something. This could be
as simple as Mr Jones getting his neighbour's son to mow his lawn
on Saturdays. There's a handshake accountability and
responsibility, but no institution or legal implication.
We use Accountability as a 'job' description. So Microsoft can have ones like 'head of Azure deployment', and Mr Jones can have his private 'lawn-mowing job' for any enterprising kid in the area.
...
Somewhat confusingly the Capabilities paper (Merrell et al) treats Capability as a Disposition, but many capabilities of real world interest are gained by education, and could be lost by lack of practice, forgetting etc.
Many BFO dispositions are just like thatTB: I am mystified by this, since learned capabilities don't have a material basis
neurology
TB: I was hoping we would not go there...
we certainly have many such innate capabilities -- e.g. we have the innate capability to learn some language or otherbut then our parents, peers, etc., have to work hard to change our neurology so that we have the capability to speak a specific language
TB: I can agree that we could regard general human
abilities to learn, remember, construct stories etc as being
materially based in having human brains. But when we get to
specfic capabilities like becoming a tennis player or learning how
to play bridge, I would suggest that any causal chain from
neuro-physiology of an individual to their prowess on the court or
at a bridge club is tenuous at best. Put another way, to consider
such learned (well or badly) abilities as dispositions of the same
sort as the disposition of Americium 241 to decay via alpha
particle emission at a certain rate seems a serious stretch to me.
The latter happens and is guaranteed to, and we have a reasonable
explanation as to why, grounded in physics - there is a chain of
causality with explanation available at every point. The same with
the effect of Hashimoto's disease on the human body or diabetes.
But there would only be a general and pretty tenuous sketch from
how Margaret Smith plays a certain style of tennis back to the
state of her brain. Some days she doesn't feel like playing
tennis, but does, other days can't when she wants to, other days
plays distractedly while thinking about a sick relative; and so
on. She might not play tennis even when the opportunity is
perfect, for whatever reason. Margaret might practice a lot and
get better (or not). Overall it seems difficult to argue that
Margaret's tennis playing 'capability' is deterministically
grounded in her neurophysiology. To even have a chance at showing
that would require a neural map at axon level, with all sorts of
feedback, modifying influences from other neural areas, and a lot
of other processing that is essentially unknown to us today. The
neural network containing the tennis skill doesn't even have to
change for her actual tennis playing to change radically because
of other brain activity.
One might argue that in principle, such a causal chain
could be established, we just lack the technology / science etc
etc today. But I don't think it is achievable because in general
the chain would be so long, complicated, and non-deterministic
i.e., specific outcomes would only be very weakly predictable from
specific initial brain states. Consequently, I think treating
learned skills as being of the same general ontological class as
water feezing or nuclear decay is hard to defend.
I will be very interested to see the new version of the paper -
it seems to me it will need to address some of these questions.
Barry Smith
On 27/06/2022 16:12, Barry Smith wrote:
On Mon, Jun 27, 2022 at 10:45 AM Woland's Cat <wolan...@gmail.com> wrote:
On 27/06/2022 00:45, Barry Smith wrote:
Their role would still be realized in the way described. Many roles (and many capabilities) are never realized.TB: I have problems with this sentence, because we don't really know what 'role' means (given 3 interpretations noted below and you have added a 4th).
A role is an externally grounded realizable entity.It does not have to be realized.That is the only sense of 'role' in BFO.TB: Just to be clear, I am not criticising BFO here, only saying that the rest of the world uses 'role' in such undisciplined ways that it's a poisoned word in a precise ontology, and thus a bit annoying, because one always has to go through the explanation of 'ok, but role in BFO means xyz etc'.
In the Party model I am working on (heavily BFO influenced), I got rid of 'role' entirely, it causes too much trouble. I have opted for 'accountability' to mean something like the description of a job post, and a (party) relationship to express the actual employment relationship (if and when it occurs).
I think accountability is too narrow -- it suggests that the person in question will be held responsible.TB: That is indeed the intention, and you may be right on it being too narrow. We are not of course trying to create another BFO, so we'll need to see if it trips us up when we work through more examples...
Nevertheless, re: your comment below 'Accountability (in my reading) involves institutional mechanisms to bring to account' - our Accountability is a bit wider than that - it is simply one Party holding another accountable to do something. This could be as simple as Mr Jones getting his neighbour's son to mow his lawn on Saturdays. There's a handshake accountability and responsibility, but no institution or legal implication.
We use Accountability as a 'job' description. So Microsoft can have ones like 'head of Azure deployment', and Mr Jones can have his private 'lawn-mowing job' for any enterprising kid in the area.
...
Somewhat confusingly the Capabilities paper (Merrell et al) treats Capability as a Disposition, but many capabilities of real world interest are gained by education, and could be lost by lack of practice, forgetting etc.
Many BFO dispositions are just like thatTB: I am mystified by this, since learned capabilities don't have a material basis
neurologyTB: I was hoping we would not go there...
we certainly have many such innate capabilities -- e.g. we have the innate capability to learn some language or otherbut then our parents, peers, etc., have to work hard to change our neurology so that we have the capability to speak a specific languageTB: I can agree that we could regard general human abilities to learn, remember, construct stories etc as being materially based in having human brains. But when we get to specfic capabilities like becoming a tennis player or learning how to play bridge, I would suggest that any causal chain from neuro-physiology of an individual to their prowess on the court or at a bridge club is tenuous at best.
Put another way, to consider such learned (well or badly) abilities as dispositions of the same sort as the disposition of Americium 241 to decay via alpha particle emission at a certain rate seems a serious stretch to me.
The latter happens and is guaranteed to, and we have a reasonable explanation as to why, grounded in physics - there is a chain of causality with explanation available at every point. The same with the effect of Hashimoto's disease on the human body or diabetes.
But there would only be a general and pretty tenuous sketch from how Margaret Smith plays a certain style of tennis back to the state of her brain.
Some days she doesn't feel like playing tennis, but does, other days can't when she wants to, other days plays distractedly while thinking about a sick relative; and so on. She might not play tennis even when the opportunity is perfect, for whatever reason. Margaret might practice a lot and get better (or not). Overall it seems difficult to argue that Margaret's tennis playing 'capability' is deterministically grounded in her neurophysiology.
To even have a chance at showing that would require a neural map at axon level, with all sorts of feedback, modifying influences from other neural areas, and a lot of other processing that is essentially unknown to us today.
The neural network containing the tennis skill doesn't even have to change for her actual tennis playing to change radically because of other brain activity.
One might argue that in principle, such a causal chain could be established, we just lack the technology / science etc etc today. But I don't think it is achievable because in general the chain would be so long, complicated, and non-deterministic i.e., specific outcomes would only be very weakly predictable from specific initial brain states.
I will be very interested to see the new version of the paper - it seems to me it will need to address some of these questions.
--
You received this message because you are subscribed to the Google Groups "BFO Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bfo-discuss...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bfo-discuss/41532fce-0ea9-fcf2-2c01-79a78cc0f343%40gmail.com.
Woland's Cat
The theory of 'grounding', which is a big part of philosophy, today, has a large subset of its proponents which see that as being entirely trivial. But delete the 'deterministically' -- no one said anything about determinism.Incidentally, our new book is in no small part about why we cannot know (much) about how the brain does these thingsAnd an AI system couldn't know either
Having discussed it a bit with Jobst some time ago, I think I am likely to agree with just about everything in the book (which I see I can now order), and it is pretty much for these sorts of reasons that I don't think learned skills can be BFO dispositions, because the latter surely implies a materially deterministic causal chain between physical state / make-up (being made of glass) and a behaviour we call a disposition (shattering when dropped).
If there is no reliable causal linking between physical makeup
and consequent behaviour, how do we know we even have a
'disposition' in BFO's sense? I think you are arguing something
like: any knowledge-based capability is a disposition because it
is a result of (neuro-)physiological state (what else could it
be?) in the general sense, even if we are quite sure we have no
hope of knowing the state, or the causal chain that gets from
state to behaviour. I actually think it's worse than that: I think
if we could know the full brain state at some time T0, we would
still have no hope of predicting behaviour for any more than a few
seconds after T0.
The problem for me at least is that I see disposition more like Chemero as quoted in the (older version of the) paper - a behaviour that is bound to occur, if the disposition for it is there, due to deterministic causal relationships.
I am also not sure if there is any problem to have a BFO class like Skill, that is a Realisable entity but not a Disposition.
Anyway I'll desist for now - I would very much like to see the new version of the paper. The book was already on my reading list, as it happens to be a pet topic of mine ;)
Thomas Beale
Maybe. But only if 'interest' is the pseudo-teleological shorthand used by evolutionists to say things like 'a frog is perfectly designed for swimming and jumping'. A evolutionary biologist doesn't really think a frog was 'designed' any more than he thinks it 'has an interest' in (say) copulation or eating mosquitos - the frog just does those things on instinct.
I also don't consider my need for my pancreas to produce insulin as an 'interest'; it's a vital necessity.
I think a relation based on the idea of utility or usefulness would be better - 'interest' can be anything, including some agent's interest in the failure of another.
Bill Duncan
--
You received this message because you are subscribed to the Google Groups "BFO Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bfo-discuss...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bfo-discuss/63bc68f2-264d-4a6d-b9ab-b233efab7e46n%40googlegroups.com.
Bill Hogan
--
You received this message because you are subscribed to the Google Groups "BFO Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bfo-discuss...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bfo-discuss/63bc68f2-264d-4a6d-b9ab-b233efab7e46n%40googlegroups.com.
Woland's Cat
>> I also don't consider my need for my pancreas to produce insulin as an 'interest'; it's a vital necessity.
Your quibble is terminological, not ontological. For the paper makes clear that these kinds of things are covered.
Arguably the kind of 'interest' that a well-functioning pancreas constitutes is ontological because the causal chain from its functioning to my health is clear. But the problem is that the same term 'interest' is covering mind-mediated capabilities, in which case an ontological claim is much harder to make.
- thomas
Bill Hogan
--
You received this message because you are subscribed to the Google Groups "BFO Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bfo-discuss...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bfo-discuss/233dfa1c-a9b7-b88d-99b4-21057a768c79%40gmail.com.
Woland's Cat
The ontological definition and usage of 'interest' in the paper is what's at stake here. And it's obviously clear.
"I have an interest in my heart pumping blood because its function is to keep me alive, and this is so whether or not I know about the existence of my heart."
"At one extreme are the interests of organisms and groups of organisms in survival and reproduction, and all the interests that flow therefrom (in obtaining food, evading predators and so forth). I have an interest in drinking water if I am thirsty. Trees have an interest in acquiring sufficient sunlight to realize photosynthesis. A cancer patient has an interest in eliminating her cancer."
Bill
Well I think what's at stake are whether all 'capabilities' in whose realisation some agent may 'have an interest' are true dispositions - that is at least my primary question.
The terminological problem with 'interest' is that it implies subjectivity on the part of the agent. A drug addict has an interest in obtaining and consuming heroin; is the ability to do so a 'capability'? For him, it is, but his family and doctor would say otherwise. So we need to be clear on whether a subjective or objective point of view is intended. Since there can be multiple competing points of view as to whether some ability fulfills some 'interest' of some agent, how is it to be decided whether any such ability is a 'capability'?
It seems to me that a 'utility' could be more easily understood
as objective, and therefore not (too) contestable.
Bill Hogan
Bill
--
You received this message because you are subscribed to the Google Groups "BFO Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bfo-discuss...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bfo-discuss/9c3da044-312a-80b7-b5bc-36ff466f9443%40gmail.com.