Content uploaded by Carlo Burelli
Author content
All content in this area was uploaded by Carlo Burelli on Mar 23, 2020
Content may be subject to copyright.
1
Political Normativity and the Functional Autonomy of Politics
Carlo Burelli
Abstract
This paper argues for a new interpretation of the realist claim that politics is autonomous from morality
and should be assessed based on political values. First, this article defends an original normative
source: functional normativity. Secondly, it advocates a substantive functional standard: political
institutions ought to be assessed by their capacity to issue binding collective decisions. Drawing from
the so-called ‘etiological account’ in philosophy of biology, I will argue that functions yield normative
standards, which are independent from morality. E.g. a good heart is one that pumps blood well and a
good army is one that it is good at exerting military force. I then interpret realism’s naturalistic
conception of politics as an etiological function of social groups: selecting binding collective decisions
under persistent disagreement. I conclude that political institutions should be evaluated realistically
by how well they perform this task. Finally, I assess trade-offs between this functional political
normativity and other moral values. I conclude that justice, fairness, freedom, equality remain
obviously important concerns, but only so long as the basic political function is secured.
Key words: Political Realism, Moralism, Political Values, Political Normativity, Functions,
Functionalism
[Penultimate Draft: forthcoming on the European Journal of Political Theory]
2
‘the function of government is to govern. A weak government,
a government which lacks authority, fails to perform its function
and is immoral in the same sense in which a corrupt judge,
a cowardly soldier, or an ignorant teacher is immoral.
(Huntington, 2006: 28).
Introduction
The old tradition of political realism has been revived to criticize the excessive focus on moral values
in political theory (Galston, 2010). While classical realists forcefully argued against utopian thinkers,
contemporary realism rejects the ‘moralism’ (Williams, 2005: 1) that views political philosophy as
‘applied ethics’ (Geuss, 2008: 6). This moralism is usually understood as the idea that proper political
philosophy is grounded on a general theory that prescribes how persons should act towards one
another, which is then deductively applied to political problems.
Realists deem moralism flawed on a methodological level for two reasons. First, it is insufficiently
attentive to the actual characteristics of political life (e.g. self-interest, power asymmetries, party
loyalty, electoral systems…) (Horton, 2010). Secondly, moralism applies moral standards designed
under these idealized assumptions to political contexts, which are strategic (Schmidtz, 2016) and
coercive (Sangiovanni, 2008) in character. Therefore, politics is conceived in an idealistic way,
roughly as the realm where we politely exchange reasons about what justice demands. As such,
moralism’s normative recommendations fail to offer relevant guidance.
Moralism is also unsatisfactory on a substantive level, because of its sharp focus on moral ideals like
justice, fairness, freedom and equality. This sometime leads to prescriptions that miss their target in
relevant real-world cases, taking for granted essential political goods like peace (Gray, 2002; Wendt,
2013), legitimacy (Rossi, 2012; Sleat, 2014a) and order (Williams, 2005; Burelli, 2019).
Moralists retort that realists risk being ‘complacent’ towards the status quo, lacking the critical bite to
oppose undesirable social practices (Estlund, 2014). By claiming that he who ‘let go of what is done
3
for what should be done learns his ruin’ (Machiavelli, 2010: 61), realists allegedly abdicate all
normativity and merely ‘propose doing what is done’ (Rousseau, 1979: 34). If realists intend to
criticize the status quo, they need to rely on some external moral ideal to warrant critical distance
(Erman and Möller, 2015). Realists are seemingly caught in a dilemma here: either they are completely
complacent towards the status quo, or they commit the same faults they ascribe to moralists.
Yet, realists intend to be normative, i.e. to offer reasons to prefer some institutional arrangements over
others (Rossi and Sleat, 2014). Rejecting unfeasible ideals, like non-ideal theorists suggest (Valentini,
2012), is not sufficient from a realist standpoint (Sleat, 2014b). Realists contend instead that politics
is autonomous: the normative force emerges from within the political sphere, not from an external
moral domain deductively applied to political questions (Jubb and Rossi, 2015).
Moralists rebut that realists are elusive. On a methodological level, it is not yet clear how this political
normativity is distinct from moral normativity (Maynard and Worsnip, 2018). Realists resist this
conclusion (Jubb, 2019), but too often seem to settle for the weaker claim that political values are a
subset of moral values, which happens to be compatible with the constraints of politics (Sleat, 2016).
This article instead tries to show that normative claims can be genuinely internal to the political sphere
and independent from moral considerations. Drawing from philosophy of biology, I explore the
hypothesis that functions provide normative implications independently of morality. Secondly, I refer
to the literature in political realism about the nature of politics, arguing that politics carries out an
important function in large social groups. I can thus conclude that realism’s political normativity can
be understood as a functional normativity of politics.
Realism has also been criticised on a substantial level, because it is not obvious to what extent it implies
different values (Erman and Möller, 2018). In fact, many realists are broadly liberals (Finlayson, 2017),
and defend mostly the same values such as equality (Jubb, 2015), liberty (Hall, 2015), non-violence
4
(Mantena, 2012) or social justice (Philp, 2016)
1
. While these values are justified in importantly
different ways from liberal moralists, I believe that realism can only radically challenge the mainstream
view if it moves beyond methodological disputes by defending different normative desiderata (Rossi,
2016). Therefore, in the second part, the paper deploys the functional normative source to ground a
different normative standard, concluding that institutions ought to be assessed politically by their
capacity to secure binding collective decisions despite persistent disagreement.
The argument unfolds as follows. First, I introduce the etiological account and show that it provides a
naturalized, objective and selective conception of functional attributions. Second, I show how this
account is not merely descriptive, but it allows normative implications to be drawn from it. Third, I
argue that the realist view of politics can be understood in a functionalist light: the way large human
groups issue binding collective decisions.
Fourth, I conclude that this very function provides a realistic normative standard to assess institutions.
Fifth, I consider how trade-offs with other moral values should be adjudicated.
The Etiological Account of Function
While the notion of function has a history of being scorned as teleological, many philosophers of
biology tried to defend a naturalized conception. The so called etiological account, inaugurated by
Larry Wright (1976)
2
, claims that in order for something to count as a function, it must have a
disposition (a) and a feedback (b). With disposition it is meant that all Xs have a tendency to Z.
However, the innovative focus is on the feedback condition, which makes this notion of function
historical and causal (hence the label ‘etiological’). It is not enough that Xs tends to Z, but it must be
1
An exception to this are those who join together realism and critical theory, following Geuss (1981) (e.g. Prinz and Rossi, 2017;
Raekstad, 2018). Those accounts often rely on ‘epistemic normativity’ to substantiate debunking or vindicatory genealogies
(Rossi, 2019: 8), so they do not necessarily need an account of political normativity. However, insofar as vindicatory
genealogies also rely on ‘functional explanations’ (e.g. Williams, 2002: 36), the etiological account may provide new insights to
this strand of realism as well, by clarifying what functions are, and why they are necessary.
2
An extensive literature followed Wright’s account. For an overview, see for example (Allen et al., 1998; Ariew et al., 2002;
McLaughlin, 2001; Moreno and Mossio, 2015)
5
the case that Z is something that causally contributed to Xs being around. In short, the etiological
account of functions states that:
In order for Z to count as a function of X, it must be true that
(a) All Xs tend to display Z
(b) Z contribute causally to Xs existence
This phrasing might seem obscure, but it is immediately clear once it is applied to an actual case:
The function of microwaves is heating food because
(a) Microwaves tend to heat food
(b) Heating food contributed causally to the existence of microwaves
In this example, someone designed microwave ovens to heat food, and thus this feature that they
exhibit played a crucial causal role in their very existence. The etiological understanding of functions
trivially applies to human artefacts, insofar as they are designed with specific purposes in mind, which
act as causal trigger to their existence. More ambitiously, the very same definition purports to fit the
biological world as well. Pumping blood, for example, can explain why we have hearts, even if nobody
designed hearts for such purpose:
The function of the heart is pumping blood because:
(a) Hearts tend to pump blood
(b) Pumping blood contributed causally to the existence of hearts
Here again, the pumping of blood can be said to play a key causal role in the fact that hearts and
organisms with hearts are around. Hearts originated from random mutations and then spread in the
world because pumping blood contributed to the fitness of the organisms which hosted them. Millikan
(1989) and Neander (1999) connect the notion of function explicitly to natural selection, referring to
such evolutionary history of hearts. As such the etiological account provides a notion of function that
is fully naturalized, something that before Darwin was thought impossible. Natural selection satisfies
the feedback condition of an etiological function, without the retroactive causation typical of
6
teleological explanations. Retroactive causation was never a problem for artefacts, insofar as the
creator’s intention comes prior to the object’s creation. However, it was a major problem for natural
functions because, absent a divine creator, there is no cause ex ante to explain the result. The pumping
of blood causally contributes to the existence of hearts, even though before hearts came to be there
was no pumping of blood. The pumping of blood cannot really be a cause of the existence of hearts –
so goes the worry – because hearts need already to be there in order to have the pumping of blood. The
process of natural evolution, however, provides a neat way to respond to this methodological difficulty:
random mutation is the causal mechanism which explains how hearts that pump blood emerge, while
natural selection explains why they persist and spread to other organisms (Mahner and Bunge, 2001).
Wright’s account thus solves the problem of the epistemically suspicious retroactive causation of
teleological final causes.
By mapping functions to causal processes in the natural world, this account allows for a definition of
function which is conceptually independent from the use that people wish to make of objects. I may
want to use the microwave in all sorts of ways – say as a drawer for my socks. However, since it was
not created as a drawer for my socks, it does not count as a function under Wright’s definition.
Similarly, I may want to die and thus I may want my heart to stop pumping blood, but my heart does
not lose its function for this reason. It was still selected for playing that role. Contrary to the etiological
account, functions are sometime conceived as dependent on the observer’s intention (Searle, 1995),
but such instrumental definitions are not particularly interesting, insofar as the very notion becomes
arbitrary, and would run contrary to an established practice in the natural sciences.
The etiological account is thus selective: it allows us to neatly distinguish what count as a function,
from a mere accidental property. Consider the following case:
(a) All hearts tend to emit thumping sounds
(-b) Yet thumping does not explain the existence of hearts
Thus, thumping is not a function of hearts
7
As we have seen, it is not enough that hearts tend to pump blood. It must be the case that pumping the
blood contributed to the existence of hearts. The opposite is true of emitting thumping sounds which
is a mere accidental quality, rather than a function of hearts. Pumping blood was causally instrumental
in the spreading of hearts, while emitting beating sounds was not. Even though it may be difficult to
ascertain empirically whether some specific disposition is a function or not, the etiological account
provides a clear framework to provide an answer.
This etiological view has an intuitive application to artefacts and biological function, but it remains
open to question whether it may be applied to the social world. Functionalism was indeed an important
research paradigm in political science (Parsons, 1991; Luhmann, 1995; Easton, 1953). Yet many
epistemological doubts were raised against functional explanations, even by those who accepted them
in the life sciences (Elster, 1994). The problem is that functional explanations defied the explanatory
golden standard of methodological individualism, i.e. the idea that only individuals command
autonomous causal power
3
. In particular, social functions seemed epistemically suspicious because
they lack a clear causal mechanism, insofar as the collective intentionality of artefacts and the natural
selection of biological functions are hardly applicable. The problem of the missing mechanism in the
social sciences partially disappears when one considers carefully what functions explain. According
to Philip Pettit (1996: 291), for example, it does not explain the first appearance of a phenomenon, but
it’s a perfectly valid explanation of its resilience, i.e. its resistance to threatening shocks. Resilience
does not involve the mysterious retroactive causation implied by some teleological explanations. A
famous and widely used example is invisible hand explanations which precisely apply the evolution
scheme to the social world and explain seemingly purposeful effects which were not produced by
intentional design. In these cases, the mechanism at play is usually some filter or a strategic
3
The canonical reference here is Max Weber: ‘in sociological work these collectivities must be treated as solely the resultants and
modes of organization of the particular acts of individual persons, since these alone can be treated as agents in a course of
subjectively understandable action’ (Weber, 1978)
8
equilibrium, that allow only some variations to survive (Nozick, 1974: 18–22). This is not too different
from the natural world, where functions are relevant because of natural selection, but the ex-ante
generative cause is random mutations
4
.
Etiological accounts of function so far offered three advantages. They are naturalistic insofar as they
rely on a causal evolutionary story, they are objective because they do not depend on the observer’s
intentions, and they are selective, insofar as they exclude other properties. A fourth interesting feature
of this definition of function is the most significant for this paper and will occupy the next section:
etiological functions are normative, rather than merely descriptive properties.
Functional Normativity
Wright’s initial analysis of function was aimed at expelling not only divine final causes, but all
evaluations from the natural world. Contrary to others
5
, he explained why some properties count as a
function without reference to the welfare of that something. Other scholars however soon observed
that his view had a quite interesting consequences: it allowed to assign functions to things that did not
discharge them (Millikan, 1989). Suppose my microwave is broken and cannot heat food. Nonetheless,
heating food still counts as its function because both the disposition and feedback conditions still apply:
(a) it is still true that microwaves tend on average to heat food, even if this broken one does not; (b)
heating food contributed to the existence of the broken microwave, insofar as it was (badly) constructed
for that purpose (Hardcastle, 2002). Less obviously, the same applies to the biological world as well.
Suppose somebody was born with a heart that cannot pump blood at all. Nonetheless, (a) hearts still
tend in general to pump blood and (b) pumping blood still explain why this defective heart exists. Even
4
Many biologists think that natural selection applies to individuals or even genes. However, Darwin quite explicitly considered group
selection (Darwin, 2003: 537–538), and recent development in biology suggest that the scope of evolution should be expanded
to encompass groups (Nowak et al., 2010).
5
For Hempel (1965), Nagel (1977), Ruse (1971) and Elster (2003) one of the essential aspects of functional explanation was the
beneficial relation of the function bearer to its containing system.
9
if I get shot in the heart, it does not lose the function of pumping blood, only its ability to perform it
(Loddo, 2016). As Karean Neander eloquently puts it:
‘To attribute a natural function […] to something is to attribute a certain
kind normative property to the thing. That is, to attribute an evaluative
standard to it that it could fail to meet, even chronically (i.e. systematically
and consistently and even under ideal circumstances)’ (Neander, 1999: 14).
The etiological account traces a sharp line between malfunctioning (failing to discharge one’s function)
and non-functioning (failing to have a function). The eyes of a blinded eagle do not lose the function
of seeing even if they cannot perform it, because this function contributed to their previous selection
and current existence. Quite the opposite, an animal from the species ‘Mexican blind cavefish’ lost its
sight after many generations of living within dark caves, to reduce the consumption of energy-hungry
photoreceptive cells and neurons. Its residual eyes lost the very function of seeing, not just the ability
to perform it.
Etiological accounts explain this intuitive difference between malfunction and non-function. Not only
functions can be attributed to things that fail to discharge them, but they also allow to rank things as
good or bad, depending on how well they perform their function. Functional normativity in this sense
implies that:
something counts as good if it performs its function well.
Consider the simple example of a knife. Knives are designed to cut, therefor their function is cutting.
In this very basic sense, a (functionally) good knife is a knife that can cut (Thomson, 2015: 69). There
are many other qualities, one might refer to in evaluating a knife: being durable, light, well balanced,
aesthetically pleasing... However, a knife that satisfies all these other desiderata, but fails its function
of being able to cut, will hardly count as a good knife. Like there is a pluralism about moral outlooks
(Berlin, 2013b), there could be a pluralism of knives: possibly all these different evaluative
considerations cannot coexist in a single knife, and people may reasonably disagree about how to rank
10
them. Yet the ability to cut seems particularly important, almost constitutive of being a knife, precisely
because it is its function.
The example of knives elucidates how functional standards are not reducible to moral ones. Obviously,
the ability to cut well is not a moral property and is only pertinent when we evaluate knives - being
able to cut will likely be a disvalue in a microwave or in a human being.
Howerver, we do not particularly care about having good knifes. Consider instead the case of
functionally good hearts, which are much more important to us than knives. If the heart’s function is
pumping blood all over the organism, a good heart is one that performs this task well. A bad heart, one
that fails to pump blood around the body satisfactorily, is something we have reasons to fix. A bad
heart, in fact, threatens our very survival. If our heart is functionally broken, we need to resort to a
machine to discharge the function that the heart was supposed to do.
Similarly, to knives, moral considerations seem completely separated from the functional goodness of
hearts. The functional standard applies regardless of whether we want a particular person to survive.
Hitler’s heart is good if it pumps his blood well, even if we have moral reasons to want Hitler dead.
Moving to the social world, consider the case of armies. Suppose the function of armies is exerting
organized violence. A good army, therefor, in a functional sense is one that can exert organized
violence. We can acknowledge for example that the Wehrmacht in WW2 was a functionally good
army, even if we deem Nazi Germany’s goal repugnant.
The point here is not that it is impossible to evaluate an army from a moral standpoint. It is obviously
plausible, following an established literature, to claim that a morally good army is one that fights only
in just wars (ius ad bello), and one that fights reasonably justly (ius in bello).
The example however conveys two important intuitions. First, the functional ground is a genuinely
normative one, which is however completely independent from other moral considerations. Second,
the moral dimension does not seem sufficient for a full evaluation of armies. If an army fully respects
11
its moral requirements, but is incapable of exerting organized violence, it would not be a good army.
It would perhaps count as a good group of people, but hardly as a good army.
Under some ethical outlooks, the functional and moral demands may even be incompatible. Let us
assume morality requires to categorically abstain from inflicting harm against other human beings. In
this case, armies can never be morally good
6
. However, we cannot too easily conclude that there should
not be armies so long as we believe that in a competitive international environment, armies are required
to preserve both political institutions and social groups living under them. If this is the case, armies
perform an etiological function, which carries normative force.
A critic may grant that functions are evaluative, but still questions whether they are truly normative,
in the sense of giving us reasons for acting one way or the other (Maynard and Worsnip, 2018: 779–
780). The thought behind this objection is that there is a strong separation between the two: I can for
example acknowledge that some painting is better than another as a good piece of art without
committing myself to do anything about it (e.g. buying it). A similar criticism could apply to functional
evaluations: I could rank various armies from best to worse, based on how well they fight, but this
ranking gives us no reason to make our army better, absent other considerations.
While I do agree that evaluations do not give us conclusive reasons for actions, it seems strange to
claim that they do not provide any reasons at all. If I recognize that some painting is a good work of
art, at a minimum I should have a preference to look at it as opposed to not looking at it. For sure this
is not a conclusive reason. Maybe the painting is located in an expensive museum or in my ex’s house
and thus, all things considered, I prefer to stay away from it. Indeed, even in the moral domain, the
transmission between evaluations and reasons for action is not straightforward. For example,
sacrificing one’s life for others can be deemed morally good, but this is not necessarily something that
I have a conclusive reason to do (maybe it is supererogatory). In the last section of the paper, I tackle
6
Indeed, in a similar way Machiavelli claimed that a good Christian might be a good person but cannot ever be a good politician
(Berlin, 2013a: 45–47).
12
the issue of tradeoffs between functional and moral standards, which is an obvious problem when
deriving actual prescriptions from functional considerations.
A second related critique is that functional normativity may be ‘parasitic’ on morality. Functions turn
into reasons for action only by relying on some moral criteria hidden in the background (Maynard and
Worsnip, 2018: 779). What makes a ‘good army’ worth having might depend on the values or goals
that a ‘good army’ can serve. For example, we might say that having a ‘good army’ helps secure
people’s wellbeing, and only for this moral reason we should make sure that we have a good army.
In response to this objection, it is worth noting that not all functions depend on moral values in such a
way. A good knife, as we remarked before, does not have any connection to moral values whatsoever.
Additionally, even when some function seems to derive their normative source from antecedent moral
values, the extent that it does so largely depends on the context. A function may be morally acceptable
or even required in some cases, but morally unacceptable in others. An army may increase collective
welfare in some cases and decrease it under other circumstances, for example in a war of expansion.
Yet even an army that engages in a costly expansion war can be qualified as a good army in its
functional sense if it is competent at fighting.
A critic may now settle on a weaker version of the previous criticism: the end need not be moral, but
functional normativity is still merely an instrumental normativity, in the sense that it only transmits
reasons from ends to means (Kolodny, 2018). It only gives us reasons for action when the end of the
function is good. Only when having an army is good for some external reason, moral or otherwise, we
should care about having a good army. This criticism attacks the autonomy of functional normativity.
In response, it is important to note that functional normativity is not fully instrumental, because it does
not arbitrarily depend on any good a person might want to pursue, moral or otherwise. Etiological
functions, as I have argued, are attributed objectively and do not depend on the observer’s disposition.
Suppose an eccentric president only cares that his/her army looks stylish when paraded down the
13
streets, but not at all whether it can fight. Instrumental normativity would imply that he/she ought to
rank armies by the elegance of their uniforms. Functional normativity still implies that they ought to
be ranked by how well they fight, because that is its essential social function.
The reason for this is that what distinguishes a function from an accidental property is the role it plays
in the survival of the system. Armies have been selected in the past because the ability to exert
organized coercion is necessary to thrive in a conflictual world, not because of the aesthetic qualities
of their uniform. So long as one cares about the system’s survival, functions are unavoidable: doing
without something that pumps blood, or without something that exerts institutional violence is not a
real option.
The critic may insist again that such survival of the system should be qualified as a moral good, and
as such functional normativity ultimately depends on a moral good. While this is a possible move, I
do not believe it is a particularly helpful one. The risk is to conflate any normativity within morality.
That would certainly make the search for a political normativity trivially false, but at the cost of
implausibly flattening the normative landscape. Similarly, I could conflate all epistemic virtues (e.g.
clarity, coherence, truthfulness) with moral values, claiming that a good person should seek to
exemplify these traits. Yet it is useful to keep in mind that someone can be epistemically virtuous and
morally misguided. One could also try to reduce prudential normativity to individual wellbeing and
claim that it has a moral ground, but again it is helpful to illuminate the possible tension between the
two. As I will argue in the next section, I think it is interesting to analyze such cases where morality
points in one direction, but political reasons urge us towards another.
In conclusion, this section argued that functions are not a merely descriptive notion and could be
attributed to things that fail to discharge them (i.e. malfunction). Functions can additionally ground
normative standards by ranking certain things in a way that is independent from moral norms (like
knives) yet important to us (like hearts). This functional normativity might conflict with moral norms
14
and take precedence over them (like armies). I shall now try to show that viewing politics as a function,
can justify the existence of the genuinely autonomous political normativity sought by political realists.
The function of politics in realist political naturalism
Functional normativity provides a framework to make sense of realist claims that there is a normativity
internal to the political sphere (Rossi and Sleat, 2014), which does not rely on external moral criteria
(Jubb and Rossi, 2015). This path is promising because all political realists lament that mainstream
liberalism betrays a ‘desire to evade, displace, or escape from politics’ (Galston, 2010: 386) and try to
bring the notion of politics back at the centre of normative theorizing (Sleat, 2014b).
So, what is politics? Politics is often considered an ‘essentially contested concept’ (Gallie, 1955),
because it has been understood in various ways throughout political theory (Alexander, 2014) and
political science (Bartolini, 2018). Some attempted to qualify this concept by reference to its specific
means of power and coercion (Jouvenel, 2000), or more narrowly as the institutional locus of power
(Sartori, 1973). Similarly, another common approach has been to equate politics to the existence and
management of conflict (Warren, 1999). Finally, politics can be also conceptualized functionally by
its output, such as ‘who gets what, when and how’ (Lasswell, 2018), as authoritative distribution of
values (Easton, 1953) or as preference aggregation (Ostrom, 1990).
Even when we commonly think about politics, its meaning is fuzzy. There are many things we do and
want to do when we engage in politics. We may be pursuing justice, dialogically exchanging reasons,
acknowledging our mutual respect and equality, or even just expressing our identities. While some try
to identify politics with all the activities that are commonly thought as political (Leftwich, 2015), this
results in an expanded term with little explanatory power.
15
Here, I will confine myself on the realist debate on the nature of politics
7
and suggest that this can be
understood as an etiological function. Doing so allows to qualify politics as a human need, that it is
not optional, but must be discharged for social groups to survive. In this sense, the etiological
conception of politics aligns with the kind of political naturalism that realists started developing (Rossi,
2009; Cross, 2017; Szűcs, 2018).
I propose to identify politics as:
The way human groups select binding collective decisions under permanent
disagreement.
This is not just one thing, though important, that political institutions might do. Rather this can be seen
as a non-optional etiological function of political institutions, because:
(a) Political institutions tend to select binding collective decisions
(b) Selecting collective decisions plays a causal role in the existence of
political institutions
(a) is quite straightforward, but (b) may seem more dubious. The feedback condition (b) however
follows from two rather uncontroversial dispositions of human beings. Unlike humans, many animals
are solitary like cats because they do not need others to survive, while others like bees have
unbreakable collective control and they are not free to put their own interests above those of the group
(Wilson, 2013). Neither are political. On the contrary, human beings can always pursue their preferences
against the resistance of others and at the same time they need to cooperate to survive. Politics thus
arises as the way human groups solve the vital problem that emerges from (1) the unlikelihood of
individual survival and the subsequent need for cooperation and (2) the instability of spontaneous
7
Although it is worth noting that the realist conception of politics is quite close to the view defended by functionalist political
scientists. For example, David Easton famously conceived politics as authoritative allocation of values: ‘without the provision
for some means of deciding among competing claims to limited values, society would be rent by constant strife […] Every
society provides some mechanisms, however rudimentary they may be, for authoritatively resolving differences about the ends
that are to be pursued, that is, for deciding who is to get what there is of the desirable things. An authoritative allocation of
values is unavoidable’ (Easton, 1953: 137)
16
cooperation. I will address both in turn. While these theses may be difficult to definitively prove
empirically, I will refer to recent work in anthropology and political science that goes in this direction.
First, cooperation not only offers many advantages (Heath, 2006), but it is a vital problem since early
human societies
8
(Bowles and Gintis, 2013; Henrich and Henrich, 2007). Individuals are not self-
sufficient and almost always form strong social ties with others. Humans are thus social animals not
in the moral sense that they care (or ought to care) about fellow human beings, but more fundamentally
in the material sense that they depend on each other to survive.
Indeed a large part of our evolutionary success hinged on our ability to cooperate, to the point that our
ancestors Homo Sapiens used their social aptitude to build complex social structures and take over
competing human species like Neanderthal, who had larger brains and significantly stronger muscle
structure (Harari, 2015). Humans are in fact neither physically well-endowed, nor smart enough to
thrive on their own. For as much as we like to praise our individual intelligence, our progress is mostly
social. Imagine if Albert Einstein were grown up in the wild, without the scientific knowledge
accumulated by our specie before him. Even such a gifted individual would struggle alone to invent
something as rudimentary as the bow, which required our equally smart ancestors thousands of years
of iterations to develop (Brown et al., 2012). This point should not be very controversial for political
philosophers, since the need for cooperation is one of the oldest and most basic assumptions of the
discipline (Aristotle, 1998: 4–5).
Second, while cooperation may be a natural need, it is not one that is naturally satisfied. Any group
will face recurrent challenges, both due to external events, and from internal conflicts. In front of such
challenges, there will be different claims about what the group should do, how it should be organised
8
Some may wonder whether politics is a natural function like the heart’s or an artificial one like the knife’s. This problem can be
bypassed by etiological accounts because both give rise to functional normativity. Moreover, sometime the distinction is not
clear: an artificial device may be used to help performing a natural function. For example, a pacemaker helps the heart in
pumping blood.
17
and how the benefits and burdens of cooperation ought to be distributed. Suppose a rival group
occupies a key resource and threaten to slay our own group unless we leave. We may disagree on
whether to resist or move away, and even then, on how to fight or where to go (and a variety of other
sub-questions). Unless these urgent questions are settled in a binding way, our group will dissolve. We
cannot simply suspend the judgment on these questions or leave them open to debate until we are all
in agreement. Such existential emergencies may seem rare nowadays, but only because they are
partially pre-empted by well-functioning political institutions (and some still occur: hurricanes,
earthquakes, plagues, economic crisis...).
This need for cooperation turns into a natural need for politics. In order to stably cooperate, human
groups need rules that establish what the group should collectively do (Rodríguez-Alcázar, 2017: 738).
Geuss identifies this wider conception of politics as ‘any form of coordinating action regardless of the
means used to achieve this coordination’ (Geuss, 2014: 149–150). In this sense it can apply to private
contexts, e.g. we can talk about the politics of a chess club with respect to the way it sanctions
collective decisions. However, this is not good analogy for politics because chess clubs are already
‘embedded in a wider, effective political association’ (Geuss, 2001: 17), which stabilizes cooperation
and coercively defuses violent conflicts. In short, collective decisions in a chess club, while significant,
are never of vital importance as they are for political societies. As a consequence, private associations
have no need to settle collective decisions in a binding way. If my chess club wants to organize a pizza
night, and I really dislike pizza, I am free to opt out and stay at home. I may not as easily avoid political
decisions I disagree with (say banning abortion or declaring a war). I will be forced to comply with
them. Furthermore, if my chess club decides to ban the special rule of castling, I am free to leave that
association and join another, or even establish a new one myself. Conversely, the costs of leaving a
political institution are very high, because individuals cannot survive alone and, under modern
conditions, there is no space outside the frontiers of political institutions.
18
For this reason, many realists associated politics not just with collective decision making, but with its
specific means of coercion. Geuss for example qualifies a narrower conception of politics as
coordination of social action which ‘makes use of at least the threat of recourse to coercion, force, or
violence, and that there is some appeal to systematic forms of legitimation’ (Geuss, 2014: 150).
Realists indeed often think that coercion is a key component in the way collective decisions are made
binding (Geuss, 2014: 150; Sleat, 2014a: 329).
Coercion, however, is not important per se, but only as a widely effective mechanism to make
collective decisions binding on everyone, particularly to those who disagree with them and would be
inclined to opt out (Bartolini, 2018; Sartori, 1973: 121). Politics, so conceived, is not about the
‘monopoly of violence’ (Weber, 2004: 33), but about the monopoly of binding collective decision
making (Schmitt, 2007).
Politics so conceived is a natural human need, insofar as it is essential for the survival of any social
group. Politics as such could still theoretically be refused, or sacrificed to some other goal, but this is
not normally the case. In this sense, politics is a need just like feeding is a need. Surely one could
refuse to feed, and indeed some people may decide to fast themselves to death – even for political
reasons. However, this does not put in question that feeding is something all human beings normally
need.
Critics of realism sometime argue that such definitions tend to be arbitrary (Erman and Möller, 2015),
and result in their opponents being ‘defined out’ of political philosophy (Estlund, 2014: 131). Viewing
politics in such functional terms is less vulnerable to this criticism because, while there are many things
that we might legitimately want to do through politics, the etiological account provide sound criteria
to single out what the function of politics truly is, even if people disagree about it. Suppose, for
example, that I claim that the function of politics is instantiating justice as fairness. Although this is
what I want politics to do, it is not its function. And we know this because it is plausible to claim that
instantiating justice as fairness is not something that all political institutions tend to do (first condition),
19
but more importantly it is not something that played a causal role in the existence of political
institutions (second condition).
While my argument is quite close to the conception of politics often defended by many realists (e.g.
Sleat, 2014b), it is worth noting that the role this plays in my argument is quite different. It is not a
filter to select among the possible moral values, the ones that are compatible with the realities of
politics (Sleat, 2016). On the contrary, it views the very function of politics as its own autonomous
normative standard. In this I believe my argument more explicitly secures the autonomy of politics
from morality.
This section crucially presented politics as an etiological function of political institutions, that
discharges a crucial natural need of individuals. Pumping blood is the function of heart, which
performs a crucial role in individual survival. Similarly, selecting binding collective decisions is the
function of political institutions, without which human groups would dissolve.
Political Normativity and its Failures
If functions yield normative standards, and if politics can be seen as a function, the political normativity
realists advocate may be originally re-interpreted as a functional normativity of politics. Substantively,
this means that good political institutions are those that can select and implement collective decisions
within a group providing its persistence over time, regardless of their fairness, freedom or equality.
While, many realists often defend equality (Jubb, 2015), liberty (Hall, 2015), non-violence (Mantena,
2012) or social justice (Philp, 2016), my view turns the very function of politics into an autonomous
normative standard: the ability to select and implement collective decisions. In so doing, this account
more explicitly vindicates the autonomy of politics from morality.
Imagining a society which lacks the ability to select binding collective decision is a useful way to
visualize why it is a desirable feature, to the point of trumping other moral considerations that might
conflict with it. Such a society would exhibit only a semblance of stability, yet it would be incapable
20
to change its rules and adapt to new conditions. Such a situation could be defined ‘sophisticated state
failure’ (Techau, 2016). Conventional state failure is the breakdown of the state in his Weberian nature,
as ‘the form of human community that (successfully) lays claim to the monopoly of legitimate physical
violence within a particular territory’ (Weber, 2004: 33). In such a situation, civil war rises because
law enforcement breaks down - a famous example being Somalia. A sophisticated state failure is
different: the state maintains the monopoly of force and keeps enforcing its laws, yet his ability to
revise the rules is eroded, and consequently its capacity to cope with new contingencies diminishes.
Having no ability to select and implement collective decision seriously hamper the long-term stability
of a social group. It is not just the ability to keep civil peace that is at stake - this is granted as a good
by many (Wendt, 2016) - but the reproducibility of a system that ensures such peace through time,
which depends on its ability to select and implement collective decisions.
Perhaps the best current example is the European Union. As a polity in the making, it comes close to
this ideal type, because it has not yet developed robust political institutions to select and implement
collective decisions (Dyson, 2013). This political flaw leads to a structural failure in responding to the
many crises it faces, and thus generates a destructive spiral which threatens its own institutional
survival. The problem is not so much the EU’s ‘democratic deficit’
9
(Majone, 1998) nor its ‘justice
deficit’ (Kochenov et al., 2015), but its political deficit, i.e. the inability to settle controversial matters
in a binding way. The problem is that the EU lacks supranational executive discretion, and this hampers
its ability to effectively counteract cyclical existential emergencies (Ferrera and Burelli, 2019). This
institution fails to comply with the outlined political normativity and is thus deficient. By not being
able to select and follow through common decision, it fails to adapt to shocks and may ultimately
9
To be fair, the EU’s lack of democratic accountability is also criticised by some realists, who claim that some kind of popular
sovereignty is necessary to establish its legitimacy (Beetz and Rossi, 2017). This is consistent with Williams’s idea of the ‘basic
legitimation demand’: political rule should make sense to those subject to it, in order to be different from mere domination
(Williams, 2005). My point is compatible, but different. To remain within Williams’s framework, my functional argument
suggests that the EU has a prior problem, insofar as it lacks the ability to answer the ‘first political question’.
21
unravel. Again, while it may exhibit a semblance of stability, when crises are absent, it is evidently
vulnerable to changing circumstances.
Political normativity also clearly matters when we are adjudicating between different institutional
reforms. For example, if we are debating the desirability of some electoral law, moral reasons might
point towards proportional representation (suppose this minimizes domination over minorities), while
political reasons may favour first-past-the-post model (see Lijphart, 1985). Moreover, this standard
shows that practices like filibustering are politically bad, and in general excessive numbers of ‘veto
players’ (Tsebelis, 2011) could pose important political problems.
It is worth noting that political normativity so construed requires significant changes to the system. As
such, it is different from mere stability and it is not vulnerable to the common objection against political
realism of being biased towards the status quo and against ambitious ideals (Erman and Möller, 2018;
Finlayson, 2017). Indeed, quite the opposite is true, as the emphasis on the ability to select binding
collective decisions increases the possibility to reform any political system, and thus if anything is
biased towards change rather than conservatism.
Summing up, this section draws the conclusion from the previous two, arguing that political
normativity can be grounded as a kind of functional normativity, and tries to portrait what a political
failure would look like.
Trade-Offs between Political and Moral Normativity
In the previous section, I concluded that we should care about how well our political institutions can
issue binding collective decisions. But should we care even at the price of other moral reasons, which
might favour less functional institutions? In this section, I argue that when trade-offs with moral values
are required, functional considerations should be given due importance, even if they are not by
themselves decisive.
22
First of all, let me note that political normativity needs not conflict with moral values. If there is room
to improve on other moral standards without losing functionality, then it is rather unproblematic to
conclude that we ought to do so. Democracy is a good example because it is not only based on the
moral ideal of political equality (Christiano, 2008), but it also involves an array of institutional
mechanisms designed to issue binding collective decisions with a high degree of compliance. For
example, in a functionalist outlook elections are a clear means of mobilizing support for the system
and its leader (Parsons, 1959).
Political normativity in fact is not only compatible with democracy, but might even require it. Albeit
this is a complicated empirical question, many political scientists acknowledge that democracies are
among the most effective political regimes at implementing collective decisions (Cheibub et al., 1996;
Przeworski, 1991). Two elements are usually emphasized as significant. First, democracies
institutionalize protests and demands for change within the political system rather than letting them
build up into potential violent revolt (Hirschman, 1970). When protest movement can organize and
voice their dissent politically from within, the potential for peaceful change raises and the interest in
violent insurrection dims. This process strengthens the ability to provide binding collective decisions,
because actors try to contest internally for power instead of publicly challenging the constituted
authority from the outside. Secondly, democracies have a procedure for institutionalized succession,
which is the greatest factor of political instability in authoritarian regimes. Electoral rules provide
shared guidelines to substitute leaders, and thus stabilize previously issued binding collective
decisions. When Kennedy was shot, his vice-president took charge until the next election selected a
new president. On the contrary, when Alexander the Great died, his generals divided the kingdom and
waged war against each other. A full argument to inquire whether democracy is the best institution to
secure the functional political standard goes far beyond the scope of this paper. However, it is still
important to highlight here that democracy is not necessarily at odds with functional normativity.
23
In certain cases, however, political normativity may turn out to be irreconcilable with certain moral
values. A key practical question therefore is how to assess trade-offs between these dimensions. It is
unfortunately impossible to give an abstract argument to assess trade-offs between political
normativity and other moral values. This question ultimately hinges on too many contextual
assumptions, like what moral values are at stake in this instance and how much they would need to be
sacrificed to ensure full functionality.
In general, however, many political realists tend to assign a special importance to political values (Jubb
and Rossi, 2015; Sleat, 2016) and the functional reading provides a good ground to explain why. As
argued before, being able to cut is not just one of the many qualities a knife should have, but one that
it is constitutive of it. Being able to pump blood is crucial for hearts and for their organisms to thrive.
Similarly, politics discharges a vital function and is not just something we are free to opt out.
Williams’s insightful intuition is that the first political question, ‘securing of order, protection, safety,
trust, and the conditions of cooperation’ (Williams, 2005: 3), has to be answered before all other moral
concerns can be raised. Andrea Sangiovanni characterizes this priority of politics in a similar way:
‘the sphere of politics not only operates differently from other spheres of
interaction […], but also is a fundamental precondition for their pursuit. […]
Political authority is necessary because without it, distrust, insecurity, and
the desire for recognition […] will thwart any possibility of cooperation or
render it incredibly fragile’ (Sangiovanni, 2008: 157)
The thought again is that politics predates other normative questions as a condition of their possibility.
Translated in the previous functionalist account, politics discharges a crucial function and allows
human groups to persist through time. Without politics, other normative questions will at best be
relevant only for a short time.
Despite this special importance of politics, it would be a mistake to go too far and try to defend an
absolute lexical priority of political normativity. Such interpretation easily looks implausible whenever
a marginal sacrifice of political performance is rewarded with enormous gains on one or more moral
dimensions. In fact, such view would leave political normativity insufficiently critical of domination
24
and would leave realism at risk of defending an excessively pessimistic outlook, ‘placing several of
today’s most suppressive regimes above the threshold’ (Erman and Möller, 2018). Most realists do
agree that ‘the core challenge of politics is to overcome anarchy without embracing tyranny’ (Galston,
2010: 391), and that we should be mindful of the risks of a ‘reduction of politics to violence by making
the de jure right to rule equivalent to the de facto ability to do so’ (Sleat, 2014a: 315). Williams in
particular was adamant that politics is not equivalent to mere domination, which is indeed the core
problem that politics aims to solve (Williams, 2005).
The best way to capture this realist intuition about the crucial importance of politics, without ending
up with an implausible lexical ordering, is to view the functional dimension as a range property: as
long as the ability of a social group to determine and implement collective decisions is satisfied to a
minimum, it does not categorically prevail over other considerations. On the contrary, when this ability
falls below the threshold on which collective survival depends, all other moral concerns become
irrelevant. The functional standard does not thus imply that other normative considerations are
irrelevant. Indeed, questions of justice, equality, liberty, democracy remain important. Yet they are
only pertinent once the functional dimension is satisfied at least to a minimum, and only so long as
they do not undermine it. Take the case of armies, we want them to respect some rules of combat, but
only once we have ascertained that they are able to fight. We may be able to sacrifice some of their
combat effectiveness for moral reasons (say by banning the use of cluster bombs), but if they are
completely unable to play their functional role of defending the polity, it makes no sense to worry
about their moral inclinations. Similarly, justice is a quality of political institutions, but if some
conception of justice were to lead to systemic failure in selecting and executing common decisions
then it would defy the proposed functional normativity. Suppose, for example, that ideal institutions
were to require unanimous consent or absence of institutional coercion. Such requirements will likely
make for bad political institutions, because they would negate their ability to issue binding collective
decisions.
25
Conclusion
This paper aimed to contribute to the recent debate about realism and its distinctively political
normativity.
My first goal was to defend a new source of normativity. I maintained that political normativity is a
specific kind of functional normativity, by arguing that functions yield normative standards and that
politics can be understood as a function of large human groups.
My second goal was to outline a genuinely political standard, independent from moral concerns like
justice, freedom, and equality. For this reason, I argued that the specific function of political institution
is to issue and implement binding collective decisions, under conditions of disagreement about what
these should be like. Without this function, human groups cannot cope with external and internal
existential shocks.
The paper suggested that the functional standard does not necessarily displace other moral concerns.
Democracies for example are very functional at securing collective decisions, because they internalize
protest and allow orderly transitions of power. When moral and political normativity do conflict, the
functional dimension takes overwhelming priority below a minimum threshold of political survival
and allow for trade-offs with other moral values above it.
26
References
Alexander J (2014) Notes Towards A Definition of Politics. Philosophy 89(2): 273–300. DOI:
10.1017/S0031819113000855.
Allen C, Lauder GV and Bekoff M (1998) Nature’s Purposes: Analyses of Function and Design in Biology.
Cambridge, Mass: Bradford Books.
Ariew A, Cummins R and Perlman M (2002) Functions: New Essays in the Philosophy of Psychology and
Biology. New edizione. Oxford: Clarendon Press.
Aristotle (1998) Politics. Indianapolis, Ind.: Hackett Publishing Company, Inc.
Bartolini S (2018) The Political. London; New York: ECPR Press.
Beetz JP and Rossi E (2017) The EU’s democratic deficit in a realist key: multilateral governance, popular
sovereignty and critical responsiveness. Transnational Legal Theory: 1–20.
Berlin I (2013a) The Originality of Machiavelli. In: Hardy H (ed.) Against the Current: Essays in the History
of Ideas. Princeton: Princeton University Press, pp. 33–100.
Berlin I (2013b) The pursuit of the ideal. In: The Crooked Timber of Humanity: Chapters in the History of Ideas.
Princeton: Princeton University Press, pp. 1–20.
Bowles S and Gintis H (2013) A Cooperative Species: Human Reciprocity and Its Evolution. Princeton
University Press.
Brown KS, Marean CW, Jacobs Z, et al. (2012) An early and enduring advanced technology originating 71,000
years ago in South Africa. Nature 491(7425): 590–593. DOI: 10.1038/nature11660.
Burelli C (2019) A Realistic Conception of Politics: Conflict, Order and Political Realism. Critical Review of
International Social and Political Philosophy: 1–23.
Cheibub JA, Przeworski A, Neto L, et al. (1996) What Makes Democracies Endure? Journal of Democracy
7(1): 39–55. DOI: 10.1353/jod.1996.0016.
Christiano T (2008) The Constitution of Equality: Democratic Authority and Its Limits. Oxford: Oxford
University Press.
Cross B (2017) Naturalist Political Realism and the First Political Question. Ratio. Available at:
http://onlinelibrary.wiley.com/doi/10.1111/rati.12165/full.
Darwin C (2003) On the Origin of Species. Broadview Press.
Dyson K (2013) Sworn to grim necessity? Imperfections of European economic governance, normative political
theory, and supreme emergency. Journal of European Integration 35(3): 207–222.
Easton D (1953) The Political System. New York , NY: Alfred A. Knopf.
Elster J (1994) Functional Explanation: In Social Science. In: Martin M and McIntyre LC (eds) Readings in the
Philosophy of Social Science. London: MIT Press, pp. 403–414.
Elster J, Matravers D and Pike J (2003) Marxism, functionalism, and game theory. In: Theory and Society.
Routledge, pp. 453–473.
Erman E and Möller N (2015) Political Legitimacy in the Real Normative World: The Priority of Morality and
the Autonomy of the Political. British Journal of Political Science 45(1): 215–233. DOI:
10.1017/S0007123413000148.
Erman E and Möller N (2018) Political Legitimacy for Our World: Where Is Political Realism Going? The
Journal of Politics 80(2): 525–538.
Estlund D (2014) Utopophobia. Philosophy & Public Affairs 42(2): 113–134. DOI: 10.1111/papa.12031.
Ferrera M and Burelli C (2019) Cross-national Solidarity and Political Sustainability in the EU after the crisis.
Journal of Common Market Studies Forthcoming. Rethinking the European Social Market Economy.
Finlayson L (2017) With radicals like these, who needs conservatives? Doom, gloom, and realism in political
theory. European Journal of Political Theory 16(3): 264–282. DOI: 1474885114568815.
Gallie WB (1955) Essentially contested concepts. In: Proceedings of the Aristotelian society, 1955, pp. 167–
198. JSTOR.
Galston WA (2010) Realism in political theory. European journal of political theory 9(4): 385–411.
Geuss R (1981) The Idea of a Critical Theory: Habermas and the Frankfurt School. Cambridge: Cambridge
University Press.
Geuss R (2001) History and Illusion in Politics. Cambridge University Press.
Geuss R (2008) Philosophy and Real Politics. Princeton: Princeton University Press.
Geuss R (2014) Politics and Architecture. In: A World without Why. Princeton; London: Princeton University
Press, pp. 144–162.
27
Gray J (2002) Two Faces of Liberalism. New York: The New Press.
Hall E (2015) Bernard Williams and the Basic Legitimation Demand: A Defence. Political Studies 63(2): 466–
480. DOI: 10.1111/1467-9248.12070.
Harari YN (2015) Sapiens. A brief history of humankind. London: HarperCollins.
Hardcastle VG (2002) On The Normativity of Functions. In: Ariew A, Cummins R, and Perlman M (eds)
Functions: New Essays in the Philosophy of Psychology and Biology. Oxford: Oxford University Press,
pp. 144–156.
Heath J (2006) The Benefits of Cooperation. Philosophy & Public Affairs 34(4): 313–351.
Hempel CG (1965) Aspects of Scientific Explanation. New York: Free Press.
Henrich J and Henrich N (2007) Why Humans Cooperate: A Cultural and Evolutionary Explanation. Oxford ;
New York: Oxford University Press.
Hirschman AO (1970) Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States.
Harvard university press.
Horton J (2010) Realism, liberal moralism and a political theory of modus vivendi. European Journal of
Political Theory 9(4): 431–448. DOI: 10.1177/1474885110374004.
Huntington SP (2006) Political Order in Changing Societies. New Haven: Yale University Press.
Jouvenel B de (2000) The Pure Theory of Politics. Indianapolis: Liberty Fund, Inc.
Jubb R (2015) The Real Value of Equality. The Journal of Politics 77(3): 679–691. DOI: 10.1086/681262.
Jubb R (2019) On What a Distinctively Political Normativity Is. Political Studies Review 17(4).
Jubb R and Rossi E (2015) Political Norms and Moral Values. Journal of Philosophical Research 40: 455–458.
DOI: 10.5840/jpr201511539.
Kochenov D, Búrca G de and Williams A (eds) (2015) Europe’s Justice Deficit? Bloomsbury Publishing.
Kolodny N (2018) Instrumental Reasons. The Oxford Handbook of Reasons and Normativity. DOI:
10.1093/oxfordhb/9780199657889.013.32.
Lasswell HD (2018) Politics: Who Gets What, When, How. Pickle Partners Publishing.
Leftwich A (2015) What Is Politics: The Activity and Its Study. John Wiley & Sons.
Lijphart A (1985) Non-majoritarian democracy: a comparison of federal and consociational theories. Publius:
the journal of federalism 15(2): 3–15.
Loddo OG (2016) Assigning functions: Agency and normativity. Sistemi intelligenti 28(2–3): 293–318.
Luhmann N (1995) Social Systems. Stanford: Stanford University Press.
Machiavelli N (2010) The Prince. Chicago: University of Chicago Press.
Mahner M and Bunge M (2001) Function and functionalism: A synthetic perspective. Philosophy of Science
68(1): 75–94.
Majone G (1998) Europe’s ‘democratic deficit’: The question of standards. European law journal 4(1): 5–28.
Mantena K (2012) Another Realism: The Politics of Gandhian Nonviolence. American Political Science Review
106(2): 455–470. DOI: 10.1017/S000305541200010X.
Maynard J and Worsnip A (2018) Is There a Distinctively Political Normativity? Ethics 128(4): 756–787. DOI:
10.1086/697449.
McLaughlin P (2001) What Functions Explain: Functional Explanation and Self-Reproducing Systems.
Cambridge ; New York: Cambridge University Press.
Millikan RG (1989) In defense of proper functions. Philosophy of science 56(2): 288–302.
Moreno A and Mossio M (2015) Biological Autonomy: A Philosophical and Theoretical Enquiry. 2015 ed.
edizione. Dordrecht: Springer Verlag.
Nagel E (1977) Teleology revisited. Journal of Philosophy (74): 261–301.
Neander K (1999) Functions and Teleology. In: Where Biology Meets Psychology: Philosophical Essays.
Cambridge, MA: MIT Press, pp. 3–26.
Nowak MA, Tarnita CE and Wilson EO (2010) The evolution of eusociality. Nature 466(7310): 1057–1062.
Nozick R (1974) Anarchy, State, and Utopia. Oxford: Blackwell.
Ostrom E (1990) Governing the Commons: The Evolution of Institutions for Collective Action. 1st edition.
Cambridge ; New York: Cambridge University Press.
Parsons T (1959) Voting and the Equilibrium of the American Political System. In: Burdick E and Broadbeck
AJ (eds) American Voting Behavior. Glencoe, Ill.: Free Press, pp. 80–120.
Parsons T (1991) The Social System. London: Routledge.
Pettit P (1996) Functional explanation and virtual selection. The British journal for the philosophy of science
47(2): 291–302.
28
Philp M (2016) Justice, Realism, and Family Care for the Aged. Social Philosophy and Policy 33(1–2): 413–
433. DOI: 10.1017/S0265052516000200.
Prinz J and Rossi E (2017) Political realism as ideology critique. Critical Review of International Social and
Political Philosophy 20(3): 348–365. DOI: 10.1080/13698230.2017.1293908.
Przeworski A (1991) Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin
America. Cambridge University Press.
Raekstad P (2018) Realism, utopianism, and radical values. European Journal of Philosophy 26(1): 145–168.
Rodríguez-Alcázar J (2017) Beyond Realism and Moralism: A Defense of Political Minimalism.
Metaphilosophy 48(5): 727–744.
Rossi E (2009) Reality and Imagination in Political Theory and Practice: On Raymond Geuss’ Realism.
European Journal of Political Theory 9(4): 204–212.
Rossi E (2012) Justice, legitimacy and (normative) authority for political realists. Critical Review of
International Social and Political Philosophy 15(2): 149–164. DOI: 10.1080/13698230.2012.651016.
Rossi E (2016) Can Realism Move Beyond a Methodenstreit? Political Theory 44(3): 410–420. DOI:
10.1177/0090591715621507.
Rossi E (2019) Being Realistic and Demanding the Impossible. Constellations. Available at: 10.1111/1467-
8675.12446 (accessed 13 January 2015).
Rossi E and Sleat M (2014) Realism in Normative Political Theory. Philosophy Compass 9(10): 689–701.
Rousseau J-J (1979) Emile: Or, On Education. New York: Basic Books.
Ruse ME (1971) Functional statements in biology. Philosophy of Science 38(1): 87–95.
Sangiovanni A (2008) Justice and the Priority of Politics to Morality*. Journal of Political Philosophy 16(2):
137–164. DOI: 10.1111/j.1467-9760.2007.00291.x.
Sartori G (1973) What is ‘Politics’. Political Theory 1(1): 5–26.
Schmidtz D (2016) A Realistic Political Ideal. Social Philosophy and Policy 33(1–2): 1–10. DOI:
10.1017/S0265052516000406.
Schmitt C (2007) The Concept of the Political: Expanded Edition. Enlarged edition. Strong TB and Strauss L
(eds). Chicago: University Of Chicago Press.
Searle JR (1995) The Construction of Social Reality. New York: Simon and Schuster.
Sleat M (2014a) Legitimacy in realist thought: Between moralism and realpolitik. Political Theory 42(3): 314–
337.
Sleat M (2014b) Realism, Liberalism and Non-ideal Theory Or, Are there Two Ways to do Realistic Political
Theory? Political Studies 64(1): 27–41. DOI: 10.1111/1467-9248.12152.
Sleat M (2016) What is a Political Value? Political Philosophy and Fidelity to Reality. Social Philosophy and
Policy 33(1–2): 252–272. DOI: 10.1017/S0265052516000285.
Szűcs ZG (2018) Aristotle’s realist regime theory. European Journal of Political Theory: 1474885118806087.
DOI: 10.1177/1474885118806087.
Techau J (2016) Sophisticated states are failing — politicians need to take risks. Available at:
https://www.ft.com/content/f519492e-022b-11e6-99cb-83242733f755 (accessed 18 June 2017).
Thomson JJ (2015) Normativity. Chicago: Open Court.
Tsebelis G (2011) Veto Players: How Political Institutions Work. Princeton University Press.
Valentini L (2012) Ideal vs. Non-ideal Theory: A Conceptual Map. Philosophy Compass 7(9): 654–664. DOI:
10.1111/j.1747-9991.2012.00500.x.
Warren ME (1999) What is Political? Journal of Theoretical Politics 11(2): 207–231. DOI:
10.1177/0951692899011002004.
Weber M (1978) Economy and Society: An Outline of Interpretive Sociology. Roth G and Wittich C (eds).
Berkley: University of California Press.
Weber M (2004) The Vocation Lectures: ‘Science as a Vocation’ & ‘Politics as a Vocation’. Owen D and Strong
TB (eds). Indianapolis: Hackett Publishing Co.
Wendt F (2013) Peace beyond compromise. Critical Review of International Social and Political Philosophy
16(4): 573–593. DOI: 10.1080/13698230.2013.810394.
Wendt F (2016) The moral standing of modus vivendi arrangements. Available at:
https://philpapers.org/rec/WENTMS.
Williams B (2002) Truth and Truthfulness: An Essay in Genealogy. Princeton: Princeton University Press.
Williams B (2005) Realism and Moralism in Political Theory. In: Hawthorn G (ed.) In the Beginning Was the
Deed: Realism and Moralism in Political Argument. Princeton: Princeton University Press, pp. 1–17.
29
Wilson EO (2013) The Social Conquest of Earth. Reprint edizione. Liveright.
Wright L (1976) Teleological Explanations. First ed edition. Berkeley: University of California Press.
