Episode post here. Many thanks to Johanna Wiedenkeller for this excellent transcription.
Matt Teichman:
Hello, and welcome to Elucidations. I’m Matt Teichman,
and with me today is Katherine Ritchie, Assistant
Professor of Philosophy at the City College of New York, CUNY.1 She’s here to
discuss social groups. Katherine Ritchie, welcome.
Katherine Ritchie:
Hi, Matt! Thanks for having me.
Matt Teichman:
Over the past decade or so, there’s been an increased
interest among philosophers in discussing the nature or metaphysics of social
groups, whether that would be teams, or corporations—or things like men,
women, white people, black people, Americans, teachers, truck drivers, etc. To a
lot of people, this seems like a non-traditional metaphysical topic. This isn’t
exactly what Kant, Aristotle, Hume, and Plato etc. were writing about when they
wrote about the grand metaphysical issues. Maybe we could begin by talking about
what metaphysics is, and then we can think about how this stuff about what
groups of people are fits into that.
Katherine Ritchie:
Sure. If we’re thinking about metaphysics, we’re
thinking about ontology. We’re thinking about questions having to do with the
nature of reality—sometimes questions about what exists. So we might ask: does
God exist? That would be an ontological question. We might ask something about
the nature of reality. What are quarks
like, supposing quarks exist?
When we’re thinking about social metaphysics, we’re thinking about questions having to do with whether certain kinds of social entities exist; are there teams, are there racial groups? Then we’re also thinking about questions having to do with the nature of these entities: if they do exist, how do they come to be? What are they like? How much can they change? Should they go out of existence? Are these good entities or not? We’re getting at those sorts of questions. So I think a lot of the questions that we’re asking about social groups are very similar to questions that have been at the center of ontology and metaphysics. They’re just applied to the social realm.
Matt Teichman:
Okay, good. Ontology is the study of what does or
doesn’t exist. Just to take a simple example, I personally do not believe that
ghosts exist. So a fancier way to say that would be to say: Matt’s ontology does
not include ghosts.
Katherine Ritchie:
Yeah, exactly. Your ontology includes what you think
exists, or a theory might have a particular ontology. So some scientific theory
might posit that certain things exist; if that is a true theory, the things it
says exist do exist. For instance, I think there are teams and committees, and I
also think there are not ghosts.
Matt Teichman:
Is this different from modern science or physics?
Couldn’t we say that back in the day, they used to think that there were ghosts,
but now we understand that this or that light effect in the sky is caused by a
chemical phenomenon? And we understand chemistry and physics pretty well now. So
is that what ontology is—waiting for our understanding of chemistry and
physics to develop? Or is it something different than that?
Katherine Ritchie:
It’s possible that chemistry or physics or other
natural sciences might give us some answers to ontological questions. But it
seems like when we’re thinking about social groups, those just aren’t the sorts
of things that at least the natural sciences are largely interested in
studying. Maybe there are some cases where we’re thinking about species. You
might think those are a kind of social group, and those are things that
biologists study.
But if we’re thinking about different kinds of social entities, some of these just don’t seem to be the kinds of things that are in the purview of natural scientific theories. Some of the questions we are also interested in, in thinking about ontology or doing metaphysics, have to do with thinking about how much something can change over time—or other sorts of features that might be central or essential to that thing’s identity. So these, again, might be questions that fall more squarely within philosophy than some of the other sciences.
Matt Teichman:
That makes a lot of sense, intuitively. It’s not like
there is some chemical property that all—and only—the members of the
Chicago Bears share. If we’re going to try to
explain what it is to be a Chicago Bear, we’re not going to talk about chemistry
stuff.
Katherine Ritchie:
Yeah, exactly! Certainly, when we get to these sorts
of groups that seem to have a kind of organizational structure, like football
teams or the Supreme
Court, it
seems like we’re really thinking about human organizations or the sorts of
societal organizations that we have created or constructed.
Matt Teichman:
Okay. Then I think the next natural question to ask here
is: why on earth would I think the Supreme Court didn’t exist? I mean, doesn’t
it obviously exist?
Katherine Ritchie:
Yes, I think it does. I think common sense certainly
tells us that the Supreme Court exists—or maybe we learned this in our
elementary school classes.
Matt Teichman:
It seems like it’s assumed by everything we hear on the
news, for example—that it exists.
Katherine Ritchie:
Yeah. Philosophers who might be inclined to think
that certain kinds of common-sensical entities don’t exist are often thinking
about the ways we could explain the same patterns (or the same outcomes, or the
same events) without positing these kinds of entities. You might think: well, it
seems like the Supreme court exists, but maybe the Supreme court is just nine
individuals. Sure, maybe those nine individuals exist, and maybe those nine
individuals do certain things—and maybe do certain things that they think are
parts of certain roles. But does that really require that there is some new kind
of entity (or some additional entity) over and above these nine individuals?
Matt Teichman:
It’s like we’re playing this game: if we had to never
talk about the Supreme Court, could we still say everything we wanted to say
about it, only talking about Supreme Court justices? Maybe there’s a fancy
little game we can play there, translating every statement about the Supreme
Court into a statement about Ruth Bader
Ginsburg and Sonia
Sotomayor and Elena
Kagan etc., rather than the Supreme
Court. Then it seems like we could do that further for each person: we don’t
even have to talk about Elena Kagan—now we can talk about the parts of Elena
Kagan. And then maybe we can talk about the cells comprising her mass, and then
we can talk about the atoms making up the cells comprising her physical mass,
etc., until we get to something that’s so sub-microscopic that it can’t be
broken down into any further parts. And maybe we can translate all talk of
anything, in some really roundabout way, into talk about fundamental physical
particles doing stuff and combining into complex systems. But the complex
systems aren’t an additional thing, over and above the physical particles.
Katherine Ritchie:
Yeah. Some philosophers have been motivated by this
idea that we should have a theory that’s simple in terms of the number of things
that exist; that has as few kinds of things existing as
possible. I’m not
particularly motivated by this, so I’m inclined to think that we can start with
the view that certain kinds of things exist. These include people as well as
subatomic particles—and things like groups as well.
If you start with a positive answer to certain kinds of ontological questions—existence questions, like “do these things exist?”—yes, it seems like they do. It seems like we have good reason to think they do. We can give clear explanations and don’t have to do all this translation work. What does the Supreme Court pick out? It picks out a particular group, and that group is varied in members across time, but it’s been around since 1789. That group. If we wanted to say the Supreme Court has been around since 1789, but we’re trying to talk only about subatomic particles—I have no idea how one would do that. At least things are going to get really complicated; maybe it’ll just turn out that it’s false that the Supreme Court has been around since 1789.
I think there are reasons to think these things do exist; I think there are reasons to think some groups might be causally relevant in ways that go beyond just subatomic particles, maybe in terms of one’s identity. Especially when you think about racial or gender or sexual orientation or ethnic groups, we might think those are really central to people’s identity. Maybe groups can actually cause things; maybe, a group can carry out some sort of action or believe something, or make a decision that doesn’t involve just the individual members making that decision, or carrying out that action.
While many philosophers are motivated to reduce things down, or eliminate certain things from their ontology, we might also start our metaphysics—or start an ontology—by thinking there are certain things that exist, and then inquire further into what those things are like. What features are central to their nature, how much change they can continue to exist through, and so on.
Matt Teichman:
It seems like the project is to come up with a theory of
what it is to be a group of people like the Supreme Court (for example). What
are the different things the Supreme Court can do? What are the ways that a
Supreme Court can change? What’s the law-like behavior governing Supreme
Court-like things? Is that the general project?
Katherine Ritchie:
Yeah, I think that’s a way of getting at the general
project. With the Supreme Court, and things like basketball teams, and chess
clubs, and baseball teams, and some sort of corporate entities, or corporate
teams, it seems like there are similarities across this class of entities. Like:
there are similarities across the class of mammals; there are differences too,
but there are some similarities. A central similarity across this class of
entities is that they’re a kind of structural entity. They have an
organizational structure that includes roles for particular individuals, roles
that can be played by different individuals at different times—and that
partially defines what these entities are. So they’re not just a bunch of
individuals, those nine people that happen to be sitting in a room wearing black
robes. Things like the Supreme Court are organized entities, or groups with an
organizational structure.
Matt Teichman:
Right; so just because you’re some people in a room
doesn’t necessarily mean you’re a group. An example of that might be: a bunch of
people waiting for the doctor, all ignoring each other, reading a magazine,
aren’t working together towards anything, as a group. They’re just some
people. But once you have something that we’re willing to dignify with the label
‘group’, there are relations that start popping up between the people. The
people in the doctor’s office aren’t related in any way; they’re just ignoring
each other. But in a group of people, there’s this structure.
Katherine Ritchie:
Yeah. I think there are different kinds of groups,
but this is getting onto something that’s exactly right. It seems like we don’t
just think that every arbitrary collection of people is a group—or we’re at
least not willing to say it’s a group. So maybe the Supreme Court is a
group. Maybe the Cleveland Cavaliers are a
group. But then think about me, Aristotle, Ruth Bader Ginsburg, and Aladdin. Is
that a group? Well, intuitively, no—it doesn’t seem like it is. What could
mark the difference between the Cleveland Cavaliers and this arbitrary set of
people I just listed off? Or what could mark the difference between women, or
African-Americans, and some arbitrary group of people that all have larger left
ears than right ears? It seems like there is a difference, and this is something
we might want a social ontology/metaphysics to give us a theory or explanation
of.
Matt Teichman:
Okay, good. So what is the difference, then, between
nine people picked by a random number generator and the Supreme Court?
Katherine Ritchie:
I think the Supreme Court and other groups like
that—not just political groups, necessarily, but teams, clubs, and things like
that as well—these are entities that have structure: you could think of them
as structured wholes, or structured entities. What does that mean? Well, it
means that there’s a certain kind of organizational setup. Just to think really
simply—the Supreme Court is pretty complicated, given its long history and its
political role—think about a club.
A club might have a president and a vice-president, maybe it also has a treasurer and a secretary, and then positions that members without specially defined roles might also play. You can think about these as positions. These positions are partially defined in terms of relations to other positions, and when some individuals start playing these positions, you have this entity that could exist through new members joining, or some members leaving. And that seems very different from just nine arbitrary people chosen at random; those individuals aren’t playing any positions. They may not be in any sort of interesting social relations. Maybe they are, but it’s certainly possible that they’re not in any sort of interesting social relations. Whereas: the justices of the Supreme Court, or the members of a basketball team, are in many pretty interesting different social relations.
Matt Teichman:
So for example, if we have a chess club, maybe the
president of the chess club stands in the ‘I can appoint you’ relation to the
treasurer.
Katherine Ritchie:
Right! We might think some of the relations that are
central to group organizational structure involve power—the power to be able
to do something, or maybe involve obligations to other members of the club,
say. I think these are the sorts of relations that would hold between the
positions in these group structures.
Matt Teichman:
That’s great. Actually, you anticipated my next
question, which was going to be: are these relations between members of a group
always, in some way, power relations?
Katherine Ritchie:
Some of the relations are definitely power
relations. I think there are relations that don’t have to do with power. Think
about baseball: the pitcher pitches, the catcher catches, the catcher might call
the pitch. Maybe that’s a kind of power relation. But then the catcher is also
going to return the ball to the pitcher, and that doesn’t seem so associated
with power—at least not this sort of power relation, like appointing someone,
or requiring someone to do something, or having the authority to fire or dismiss
someone. This seems like a different kind of relation. So I think there could be
(and are) relations in certain kinds of groups that aren’t related to power, but
many of the relations we’re especially interested in do have to do with power
hierarchy obligations, or what one is expected to do.
Matt Teichman:
Can two members of a group not be related? I was
thinking the shortstop and the outfielder, in the baseball case, have totally
separate roles.
Katherine Ritchie:
Yeah. I think certain positions or certain roles in
a structure might be more intimately related (or more closely related) than
others. I’m inclined to think that the way these structures are going to be
understood—if we were drawing them out on a piece of paper, with circles where
people might stand, and different relations between those—that all of them
will at least be related through other positions. Even if they’re not directly
related, they’re part of this organization, and this organization maybe shares
some sort of common goal or project. And even if certain roles aren’t closely
related, they’ll be related in a transitive sense, in that role A might be
directly related or closely related to role B, and B might be closely
related to role C. Maybe A and C aren’t as closely related; nevertheless,
they’re both related to B. So they are part of this larger structure that
involves various different relations.
Matt Teichman:
So just sticking with the baseball example, players who
play the positions that are far from each other on the field might still be
aware that in certain situations, what I want to do is throw the ball quickly to
my neighbor. And there, I know that so-and-so is my neighbor. And then another
player is that player’s neighbor, but not my neighbor. But in order to get the
ball to the person two hops down, I need to know about who is whose neighbor in
order to see what path the ball is going to travel through the different
players, as we’re throwing it to each other.
Katherine Ritchie:
Exactly. It seems like there is going to be a
broader sense of cooperation or interaction, and maybe a representation, by
others on the team, of other people’s roles. Certainly, on baseball teams, it
seems like players should, at least, have representations of what other players
are expected or likely to do. And that helps them play well together.
Matt Teichman:
That’s interesting: it seems like in all the cases we’ve
discussed, there are rules or conventions dictating different roles. Then the
people in the groups occupy the different roles, and maybe the roles then
require that they stand in different relations to each other. Or maybe they have
different things they’re allowed to do with each other, depending on what the
rules are, in that specific case—maybe there are the rules of baseball, and
then there are the legal rules of the Supreme Court. Does that apply generally,
or is that only for certain kinds of groups—that they’re rule and
convention-based?
Katherine Ritchie:
I think the sorts of cases we’ve been talking
about—especially teams that are parts of major leagues, and things like the
Supreme Court, or Senate, or the House of
Representatives—involve a lot of conventions and a
lot of rules. Maybe legal rules, maybe rules within the franchise or within some
larger organization. I think there are some groups that could develop in a way
that’s more organic, like a chess club coming to be, or a book club starting
up. Maybe it starts out with people sending a couple emails around and deciding
to read a book. Some roles start to become defined; they might become defined in
a way that eventually becomes conventionalized, but isn’t as explicitly
conventional, to begin with.
We’ve focused on groups that seem to be highly organized, but you might think: well, the sorts of groups that we often talk about the most are things like ethnic groups, racial groups, immigrant groups. You might think some of these don’t have defined roles for particular individuals. I’m inclined to think that that’s right, and that there’s a pretty stark difference between things like a basketball team and things like gender groups. But I think that these can also be understood in terms of larger, broader societal structural relations. Here again, I’d be less inclined to say these are based on sort of explicit conventions or rules: rather, they’re based more on patterns in society—patterns having to do with what one is supposed to do, how one is supposed to act, how one is treated, sometimes what legal rights one has, and so on.
Matt Teichman:
Maybe in some groups, the relations between the members
of the groups spring up in this more bottom-up manner, while other groups were
initially defined or conceived by reference to the relations between the
members. An example of a bottom-up type group would be a friend group where
nobody decided we were all going to be friends, but it just so happened that
because of everybody’s schedules, randomness, and who knows what, everybody
started sitting together at this lunch table. Nobody decided that so-and-so
was going to be the charismatic leader of the friend group; it’s not like there
was an election they held, where they decided that Hank over here was going to
be the charismatic leader. It just so happened that certain people started
behaving certain ways. Patterns of deference to this person sprung up
organically. And then, what do you know? You fast forward a year in advance, and
now there’s this actual friend group with a little social hierarchy in it, even
though it wasn’t formally stipulated.
Katherine Ritchie:
I think that’s right. It seems like sometimes, a
group structure might be explicitly set out: people deliberate and decide how
they want this group to be structured. Maybe this is set up through particular
intentions, so they really intend for this to be the structure of the group. And
in other cases, something might develop in a looser sort of sense—maybe not
through explicit intentions. Maybe no one intended a certain person to be the
leader of the friend group, but nevertheless, we think at a certain point that
that person clearly is the ‘central officer’ of the friend group!
Matt Teichman:
Deciding which mini marts to go hang out at, or
whatever. If I had to pick which one of these two things a gender group was, I’m
not sure what I’d say, but I guess it’s more like the bottom-up type case—like
the friend group, as opposed to the Supreme Court, or a team. But I’m not
totally sure. What do you think—is that the right question to ask about a
category like ‘men’ or ‘women’?
Katherine Ritchie:
This is a really interesting question, and it gets
at an issue that people working in feminist
philosophy, feminist
theory, and critical race
theory have thought a
lot about. People often think that gender, or race, or ethnicity is a natural
feature in reality; that maybe it’s biological or genetic. And this is something
that a lot of theorists have pushed strongly against. They have argued that
while these things might appear to be natural, they’re actually social in
nature, so they depend on different sorts of social arrangements or social
patterns.
Let’s think about the case of gender groups. When thinking about a gender group like women, one might think that this is a natural category. This is something that feminist theorists, feminist philosophers, and some critical race theorists have pushed against; they have pushed against the notion that these are just biological categories, rather than social categories.
If these are social categories, are they to be understood in terms of structure? I’m inclined to think that they are—that we can think of genders as being much broader societal structures. Individuals that are gendered as women (or identify as women) are positioned in a particular place in this larger structure. This place might be partially defined in terms of power relations to other groups. It might be defined partially in terms of what’s expected of individuals that occupy those roles. It might be partially identified in terms of how people identify and what they find really central and valuable about identifying with that sort of gender group. But I think these often will be societal structures that aren’t intentionally created in the way the Supreme Court was. So I think there is a difference in intention or explicit overt action in creating a structure of a team, versus structures having to do with economic class, or gender, or race.
Matt Teichman:
Maybe it’s a bit like the friend group case, except that
it’s a really big group of about 3.75 billion people each, or whatever it
is—assuming two genders, anyway, which of course is oversimplifying. Maybe
it’s similar to the kind of thing we were talking about with the friend group,
except with way more people and over a longer timescale. So for the length of
human recorded history, there have been groups that have been called these
things: ‘men’ and ‘women’. Maybe the scale of how long the group has been around
is also a bit overwhelming. Do you think that is a helpful way to think of some
of these categories—that they’re like the smaller groups we started off
talking about, but they have way more people in them and have existed for way
longer?
Katherine Ritchie:
I think in some sense, that’s right. Certainly it
seems like gender has existed longer than any particular friend group. There is
a difference in the length of time that these groups have existed. While I think
social groups generally are related to structure—at least the sort of central
cases of social groups we’ve been talking about are related to social
structures—I think there’s a pretty major difference between social groups
like courts, teams, committees on the one hand, and then social groups like
gender, and racial groups, and economic classes, on the other. And I think this
comes down to a distinction between what we might think of as entities that are
more object-like—object-like, like a lamp. Object-like, like a basketball
team.
Maybe that sounds a little funny. But these are entities that exist in the world. There can be more than one lamp—there is more than one lamp. There can be more than one basketball team; presumably there are many basketball teams. Whereas when you think of things like women—or if you think of a more intersectional group like working class African-American heterosexual women, you’re thinking about something that’s more kind-like. Or, at least, this something that philosophers sometimes call ‘kinds’. This presumably wouldn’t be a natural kind—a kind that is found in nature outside of societal structures—but a social kind. It’s the sort of thing that can change in members dramatically. Maybe teams can as well, but it seems to be less of a particular sort of entity. When you think of a gender category or gender group, there are however many there are, but it’s not the same as a friend group. You can talk about 20 different basketball teams playing in a tournament, but you wouldn’t talk about 20 different—
Matt Teichman:
—yeah, I think I can see the intuition here. For each
of these categories, there’s ‘one’ of them: if there’s a group of people who are
heterosexual, African-American working class women, that is one group, not 20
different ones. Whereas with a corporation, or a sports team, there’ll be more
than one.
Katherine Ritchie:
In the case of these entity-like groups, it seems
like there can be more than one. Whereas when you’re thinking about kinds of
entities—or certain social groups that seem like kinds—it’s like: there are
lamps, and that’s a kind of thing. Of course there can be a bunch of different
lamps, but there’s just one kind, at least if you get specific enough. And
similarly, you might think that some kinds of social groups, like economic
groups, gender groups, or racial groups, are like that. There’s a bunch of
individual people in the group at a particular time; whether because they
identify as that or because others identify them as that. But the
group—there’s just one of them.
Matt Teichman:
This counting thing is interesting. It’s like we count
kinds differently from the way we count—I don’t know what to call
them—non-kinds. If you were to say, how many kinds of rat are there? It seems
like the answer to that would be sub-kinds of rats. There is the Norwegian rat,
and the black rat, and the whathaveyou rat. Whereas if you ask how many
basketball teams there are, you’re not going to give sub-parts of the basketball
team as an answer to that.
Katherine Ritchie:
You could say that there are professional
basketball teams, high school basketball teams, college basketball teams and
intermural basketball teams. But usually, if you ask how many basketball teams
there are, you’re going to say how many basketball teams there are in this
division, or in the NBA, or in March
Madness,
or whatever. It seems like this counting question also relates back to issues
having to do with ontology that we were talking about a little bit ago.
When philosophers are asking whether there are certain kinds of entities, or what entities you include in your ontology—‘what’s in my ontology?’—they’re often thinking about what kinds of entities are going to show up in their ontology. They’re not going to say, ‘well, I believe in tables’ and then say, ‘I better go count how many tables I need to put in my ontology, see how many there are’—rather, they’ll say, ‘I believe in tables’, or ‘furniture’. Or ‘I believe in material objects’, and that includes tables, and chairs, and a bunch of other things. Similarly, we might think this sort of difference between particular entities, like a particular team or particular lamp, versus kinds of entities—this distinction that we’re seeing in social groups, that I think holds between different sorts of social groups—also shows up as a more general question in ontology.
Matt Teichman:
You’ve argued that there can be two different groups of
people that have all the exact same individuals in them, but they’re still two
different groups. And that this conception of a group as being a structured
whole puts you in a good place to explain what it means to say that the two
different groups can have exactly the same members. What would be an example of
that?
Katherine Ritchie:
We might think of a basketball team and a book
club. Let’s say that the Cleveland Cavaliers decide to start a book club. All of
the members of the Cavs, right now, are also members of the book club, right
now. But you might think that the basketball team isn’t identical to the book
club. When the book club meets, they’re sitting around, they’re talking about a
book; they’re not wearing basketball shoes, and even if they are, they’re not
playing basketball. When the Cavaliers meet on the court, they’re dribbling the
ball, they’re playing another team, or they’re practicing. So it seems like
they’re doing very different sorts of things. And if you have this picture of
these organized groups as structured wholes—or entities that have
structures—the difference can be captured in terms of different structures in
these two entities.
Maybe the book club has a structure that involves a president, and a vice-president—or just a bunch of members. Maybe each person plays a very similar role. And the basketball team has different sorts of positions that different individuals play. The structures are different. So even though the same individuals are playing roles in the book club and playing roles in the basketball team, the entities are distinct, because they have different structures.
If you don’t have a view on which these groups also have structural elements that are central to their identities—are really parts of what they are—you just think groups are a bunch of people. You’re not going to be able to capture this difference, because you just have one bunch of people. So if you start listing out and pointing to these individuals, they’re just ‘there’. You just have these individuals, and there’s not a good way to capture the difference between the individuals making up a basketball team and the individuals making up a book club. Nevertheless, it seems that we could have two distinct groups here: the club, and the team.
Matt Teichman:
I thought it was really interesting how—when you
introduced this example of a basketball team that happened to be made up of the
same people as a book club, but we definitely want to say they’re different
groups—I thought it was interesting how you characterized that, at first, as:
they do different stuff. The book club people meet together in order to read
books, and the basketball team meets to play basketball games. Is this idea that
a group has a structure to it consisting of relations between people—is that
just another way of expressing the idea that different groups of people do
different types of things?
Katherine Ritchie:
I think group structures do get at part of what
groups do. So if you think about the different roles (and the different
relations between roles) in a group—often, just looking at that—if you knew
nothing else besides what these roles were and how they were related to one
another, and certain features that people needed to have to play the role (or
that they needed to be people playing the role), I think you would find out
quite a bit about what the group was up to, or what the group was often aimed
at.
But groups are also parts of larger social structures. Two teams might play each other in a game. Then you’ve got a team and another team: both of these have internal structures that involve roles and relations between the roles. The game has certain rules that specify how the teams are to interact when the game ends, and who wins the game. I think that to really get the full picture of what a certain kind of group is aimed at—or what a certain kind of group is doing—the internal structure will give you part of answer to that, but then you’ll need to look outside it as well.
Matt Teichman:
Why do you think it’s helpful to come up with a good
theory of what groups of people are, and what kinds of properties they can have,
and what they can do?
Katherine Ritchie:
Considering issues in social metaphysics, or
considering what social groups are, can be important in a lot of different
ways. I think it’s interesting to get a better, clearer, understanding of what
the world is like. That’s what we do generally in philosophy, science, history,
and all sorts of other disciplines—we’re trying to figure out what the world
is like. Here is a part of the world that’s really central to our lives: what
groups we’re part of. If you think of families and our identities in terms of
our nationality, our gender, and race, and ethnicity, if you think about the
sorts of organizations we care about, or the teams we root for, these are all
really central aspects of our lives. So it’s interesting just to try to figure
out what these things are like.
But then if we’re going beyond the intrinsic interesting-ness, which at least I find, there are a lot of issues in philosophy, politics, and social justice, more generally, that a clear understanding of the nature of social groups might help to inform. We can think about issues in ethics: do groups have responsibilities? Do groups have obligations? Can groups actually carry out some action or make a decision? Do we have to respect the rights of groups—not just the individuals, but groups themselves? There are also issues having to do with knowledge, or whether groups can know things, or whether being a member of a group could affect whether someone’s testimony is believed more or less.
And then thinking about issues in say, identity politics, how should we understand identity politics? Is this something that we ought to pursue or ought not to pursue? What kinds of groups are involved in identity politics? Have we moved to the level of something more like a team, or a court, where an organized sort of entity is carrying out some sort of politics? Or are we more at the level of a kind-like group, like a gender group or an ethnic group? Metaphysical issues might be informative in thinking about whether identity politics is good, or not—or should be pursued, or not.
And then, when thinking about social justice—we don’t live in the most just world. That seems true. We want to live in a more just world. (I hope that’s true for everyone!) How do we do that? How do we make the world more just? Well, a step in that process is understanding what the world is like. And a step that’s really central to social justice projects is thinking about what social groups are like, how they came to be, how we might change them, and what features they have. I think these are all elements that could be part of a social justice project. Certainly not the entirety of a social justice project, but a preliminary step in that process.
Matt Teichman:
Yeah, I have to say I completely agree with that. It
seems to me that we’re really interested in this type of stuff. Like: what does
the fact that you’re a member of such and such group entitle you to say? If I’m
a fat person, does that entitle me to speak on behalf of fat people? Well, it’s
very complicated; who really knows? Does anybody have the right to speak on
behalf of anybody, or do we have more or less of a right to speak on behalf of
each other, depending on who we are? And what even is ‘who we are’? These are
questions that we’re all super interested in right now, in our cultural moment.
Maybe, in order to be able to have those conversations, we need to get clear
about what a group even is.
Katherine Ritchie:
Yeah! Thinking more about the nature of groups—or
at least some people spending some time thinking about the nature of
groups—is really useful. I hope it’s useful. It can help clear some space
and allow us to think in a more productive way, hopefully, about how to
understand what the world is like. And how to better shape the world to create
this more just social situation that we want to be the case.
Matt Teichman:
Katherine Ritchie, thanks so much for joining us.
Katherine Ritchie:
Thanks for having me!
Elucidations isn't set up for blog comments currently, but if you have any thoughts or questions, please feel free to reach out on Twitter!
- At the time of this transcription, Katherine Ritchie is Assistant Professor of Philosophy at the CUNY Graduate Center. [return]