Dealing with Dictators

Full text

1
Dealing with Dictators
1
Forthcoming in Journal of Political Philosophy
Chris Armstrong, University of Southampton
Introduction
Dictatorship is a great stain upon our world. Dictators treat their subjects purely as means to
their own aggrandizement, and not as agents with a right to participate in self-government.
Moreover dictators frequently, though not always, commit great harms: starting wars,
brutalizing dissidents, persecuting minorities, and squandering resources which could
otherwise be used to advance the common good.
The question of how outsiders – including outsiders living in societies committed in
word, if not always in deed, to the core principles of liberal democracy - ought to engage with
or respond to dictatorial regimes, though, is a thorny one. Difficult questions arise in the context
of overseas aid, for instance, as states and NGOs grapple with how to act in a situation where
their support might bolster repressive regimes.
2
But the focus here will be upon international
trade. In recent years, a number of scholars have argued that for liberal democracies to allow
their citizens and corporations to trade with dictatorships can also serve to entrench dictators’
positions, and even make the emergence of dictatorships more likely. To avoid such outcomes,
1
For helpful advice, I would like to thank two anonymous referees, James Christensen, Cecile Fabre,
Jonathan Havercroft, Eszter Kollar, Cara Nine, Kieran Oberman, David Owen, James Pattison, and
Patrick Tomlin, as well as audiences as the Katholieke University Leuven, University College Cork,
and the University of Oslo.
2
Jennifer Rubenstein (2015) Between Samaritans and States: The Political Ethics of Humanitarian
INGOs. Oxford: Oxford University Press, 87-114.

2
our governments ought to forbid citizens and corporations from buying goods from societies
blighted by dictatorship. Leif Wenar has recently argued that we should stop buying natural
resources from countries where governments are not accountable to citizens in even a minimal
sense.
3
Thomas Pogge has argued that our willingness to trade no-questions-asked with
dictatorships serves to protect – and even incentivise the emergence of – brutal regimes, and
as such violates a negative duty not to harm the global poor. In at least some circumstances,
we are morally obliged to break off trade – including cases where initially legitimate
governments in vendor countries slide away from democracy.
4
Shmuel Nili has claimed that
democracies should not trade with dictatorships at all, since doing so would be a hypocritical
violation of our liberal and democratic commitments.
5
These arguments all suggest that morality requires the cessation of, or at the very least
serious restrictions on, trade with dictatorships, but they specify these restrictions in different
ways. The kinds of trade which ought to be prohibited can be calibrated more broadly, or more
narrowly, on at least two scalar dimensions. We might be obliged, more narrowly, to cease all
purchases from dictators or their cliques only, or, more broadly, to cease all purchases from
societies blighted by their rule whether they are routed through the ruling elite or not. Or we
might argue for some intermediate position. And we might be obliged to cease purchases of
some kinds of goods, more narrowly, all goods, more broadly, or something in between. Nili’s
account appears to be broadly specified on both dimensions. Pogge’s and Wenar’s focus
appears to be on purchases from unrepresentative ruling elites. And their focus is relatively
3
Leif Wenar (2016) Blood Oil. Oxford: Oxford University Press, 263-336.
4
Thomas Pogge (2002) World Poverty and Human Rights. Cambridge: Polity, 152-73. Pogge’s
proposal is conditional. Democratic states ought to pass laws stating that outsiders can no longer buy
public property from them if they slide back to authoritarianism. If they pass such laws, and later slide
into authoritarianism, outsiders should consider popular consent to such sales to be withdrawn. We will
return to the issue of consent later.
5
Shmuel Nili (2011) “Conceptualizing the Curse: Two Views On Our Responsibility for the Resource
Curse,” Ethics & Global Politics 4.2: 103-124.

3
narrow in our second sense too: they concentrate on the sale of public property, rather than all
goods and services.
6
In what follows I am interested in the full range of arguments here. I will consider both
arguments for widely-applied restrictions on purchases of the kind endorsed by Nili, and
arguments for narrower restrictions on trade of the sort associated with Pogge and Wenar. In
principle it might be that restrictions which are narrower on one or both dimensions are easier
to defend – but that remains to be seen. It is also worth clarifying that the claim I am interested
in is not that we should cease trade with dictatorships when they engage in genocide, mass
atrocities, and ethnic cleansing. The claim I am interested in holds that we should cease trade
(in the specified ways) with dictatorships whether they engage in these activities or not.
Likewise, I will not be investigating the claim that we should cut off trade whenever trading
renders us beneficiaries from injustice, or complicit in exploitation. Neither concern, after all,
gives us reasons for breaking off trade with dictatorships in particular. The arguments I am
investigating, by contrast, attempt to supply particular reasons for doing just that.
My focus will be on purchases, since arguments that there are some goods – such as
weapons – which we should not sell to dictatorships have already received considerable
attention.
7
I believe that arguments for restrictions on our purchases possess some initial
plausibility. If dictatorship is a great moral bad, then we might think that we should make
strenuous efforts not to be complicit in it – and given the projects our money might serve,
providing dictators with a steady stream of revenue looks like an important example of such
complicity.
8
Some sectors of international trade might be especially pernicious in their effects.
6
Wenar’s focus in Blood Oil is still more narrow, as the book’s title suggests. Oil might indeed be an
especially pressing case, given the pathologies associated with dependence on it. But if the claim is that
what is owned by the people should not be bought from dictators, the argument should extend, like
Pogge’s, to all forms of public property.
7
See e.g. James Christensen (2019) “Arming the Outlaws: On the Moral Limits of the Arms Trade,”
Political Studies 67.1: 116-31.
8
For an analysis of various forms of complicity see Chiara Lepora and Robert Goodin (2013) On
Complicity and Compromise. Oxford: Oxford University Press, 17-30.

4
The resource trade, for instance, delivers a source of income which might insulate leaders from
citizens’ demands for political inclusion. The ready availability of resource income is
sometimes said to reduce rulers’ need to tax, and hence to engage with the demands of, ordinary
citizens.
9
There are several possible arguments for the conclusion that we are duty-bound to
stop buying from dictatorships, however. To make progress, it is helpful to distinguish between
‘instrumental’ and ‘non-instrumental’ arguments. Instrumental arguments suggest that we
ought to cease trade in order to advance valuable goals such as human rights or democracy (in
societies blighted by dictatorship). Non-instrumental arguments do not focus on outcomes.
Instead they suggest that the cessation of purchases is morally imperative insofar as it expresses
an appropriate repugnance for dictatorship and oppression, for instance, or signals due respect
for values such as property rights or consent. These arguments suggest that cessation may be
required even if it does not improve the situation of those ruled over by dictators – and even,
perhaps, if cessation makes things worse for them.
I will examine each type of argument in what follows. My overall conclusion will be that
we ought to be very cautious, in fact, about withdrawing from trade with dictatorships. When
it comes to instrumental arguments, an engagement with the extensive empirical and normative
discussions that have taken place on the topic of economic sanctions in recent years reveals
that we have serious grounds for concern that the cessation of trade with dictatorships will not
achieve its stated ends, and may make things worse – and perhaps much worse – for the people
we are trying to help. I will also suggest that while cessation may be more successful in
advancing its non-instrumental goals, the cases in which we are obliged all things considered
to cease purchases are rarer than we might suppose. Finally, I will inject a useful comparative
element into the discussion by addressing empirical work on what typically happens when we
extend and deepen, rather than withdrawing, trade with dictatorships. If the relevant empirical
9
See e.g. Michael Ross (2012) The Oil Curse. Princeton: Princeton University Press, 27-62.

5
findings are robust, they throw interesting new light on our question. In the end, I will argue,
at least in the majority of cases we are likely to confront, the withdrawal of trade with
dictatorships will not be morally required – and indeed may be impermissible. Surprisingly,
perhaps, our best response to the horror of dictatorship may turn out to be to trade more, rather
than less.
Instrumental Arguments
Terminating our trading relationship with dictatorships is a paradigmatic example of the
economic sanction. Though refusing to sell goods to dictatorships is probably the more
common strategy, refusing to buy from them is also a familiar element of many packages of
sanctions. The sanctions imposed on Iraq following the Gulf War, for example, centrally
involved constraints on its ability to sell oil – though these constraints were eventually
moderated under the ‘oil for food’ scheme. Similar embargoes have been placed on Iran and
Syria in more recent years.
10
Blocking purchases of goods such as oil, diamonds or gold
removes from dictators a major source of income – income which could otherwise fund
weapons transfers and the repressive apparatus of the state (among other things).
It is important to note that the cessation of trade our critics have in mind is distinct from
the individual consumer boycott. We might suppose that individual consumers exercise an
important freedom in refusing to buy from certain producers, or certain places – and that the
justificatory bar for such boycotts ought therefore to be set rather low.
11
But the proposals under
review remove from individual consumers this freedom to choose. On Wenar’s proposal, for
10
Erica Moret (2015) “Humanitarian Impacts of Economic Sanctions on Iran and Syria,” European
Security 24.1: 120-140.
11
For an analysis of moral limits to the individual freedom to boycott, see Waheed Hussain (2012) “Is
Ethical Consumerism an Impermissible Form of Vigilantism?” Philosophy & Public Affairs 40.2: 111-
143.

6
instance, our government should act to forbid individual consumers and corporations from
buying resources from the target country. And it should stand ready to employ coercive legal
measures in order to enforce cessation.
12
As Cecile Fabre observes, economic sanctions
interfere in economic relationships between consenting parties.
13
Indeed – since unilateral
sanctions imposed by a single country are rarely effective - states will often go further,
employing what Fabre calls ‘secondary sanctions’ in order to discourage actors in other states
from trading with the target - again with the threat of coercive measures in the background.
This kind of restriction requires a more powerful justification.
14
The justificatory bar may be
lifted higher still if, as I suggest in the final section of this paper, trade typically engenders at
least some positive outcomes which sanctions would thwart.
In practice, sanctions have often been instruments of power politics, aimed at pressuring
other states to fall in line with various foreign policy goals. But they also have objectives which
are normatively plausible on their face. In this section, I will focus on instrumental arguments
for the withdrawal of trade. Sometimes, to be sure, sanctions seek to stop mass atrocities,
genocide or the invasion of other states. I will not challenge the idea that sanctions – or indeed
the use of force – would be appropriate in such circumstances. The claim I am examining is
that we ought to cease trade with dictatorships even when they do not engage in these great
crimes. The instrumental argument I am interested in is aimed at preventing or deterring more
quotidian offences. For instance, the cessation of purchases might seek to prevent the breaches
of physical integrity rights which often occur in dictatorships, including police brutality,
arbitrary detention, and torture. Or it might seek to promote rights of political participation,
setting dictators on the path to democratic reform, or at least engendering more narrow forms
12
Wenar, Blood Oil, 283-8.
13
Cecile Fabre (2018) Economic Statecraft. Oxford: Oxford University Press, at 29.
14
Daniel Weinstock argues, relatedly, that whereas boycotts usually face a low justificatory burden,
imposing them coercively imposes a greater burden. Weinstock (2018) “Dissidents and Innocents: Hard
Cases for a Political Philosophy of Boycotts,” Journal of Applied Philosophy (earlyview).

7
of accountability. Economic sanctions also sometimes aim at regime change – though insofar
as regime change is desirable, its value will primarily lie, we might suppose, in the ensuing
gains to human rights or democracy.
Fortunately, prima facie-plausible criteria with which to assess the instrumental effects
of sanctions are not hard to come by. In recent years, there has been a small explosion of
literature on the morality of economic sanctions. For the most part, this new literature seeks to
apply concepts which are more familiar from just war theory. Just war theorists have attempted
to isolate conditions which must be in place for the instigation of lethal force to be justified.
The key thought for our purposes has been that at least some such conditions are relevant in
the case of sanctions too. To be sure, we might well believe that the justificatory standard for
economic sanctions should be lower than in the case of military force. When we stop buying
goods from dictatorships, we are not launching lethal ‘kinetic’ actions, as just war theorists
sometimes put it. We are simply withdrawing from trade – albeit with the potential use of
coercion against citizens, or third parties, who refuse to go along with cessation lurking in the
background. As Fabre puts it, ‘although we might end up killing people, we do so by interfering
with contractual economic and financial relationships. A precise philosophical justification for
sanctions must be sensitive to that distinction.’
15
But just war conditions including effectiveness and proportionality appear to be relevant
nevertheless. The standard of effectiveness concerns the probability that the goals of our
actions will be brought about by those actions; the standard of proportionality, by contrast,
requires a comparison of the morally weighted harms our actions will produce with the morally
weighted harms which would come to pass if we did not act.
16
If the justification for the
cessation of trade is going to be instrumental in the sense that it aims to improve the situation
15
Fabre, Economic Statecraft, 34.
16
See Henry Shue (2018) “Last Resort and Proportionality,” in Seth Lazar and Helen Frowe (eds) The
Oxford Handbook of the Ethics of War (Oxford: Oxford University Press), 260-76.

8
of individual subjects in dictatorships, then it would be very odd not to apply these standards
to our actions. The negative consequences our choices generate at least ought to be no greater
than the harms that would come about if we stood by and did nothing; and we must be
reasonably confident that our actions are going to achieve their stated ends.
On this basis, however, there are grounds for serious caution. Sanctions became very
popular in the years immediately following the Cold War. The United Nations Security Council
imposed economic sanctions twice between 1945 and 1990, but 11 times during the 1990s. The
US imposed sanctions on 35 countries between 1993 and 1996 alone.
17
But a backlash did not
take long to occur. By the end of the 90s it had been claimed, for instance, that sanctions had
contributed to more deaths in the decade since the Cold War than all weapons of mass
destruction throughout history.
18
At the same time, it was alleged that resort to sanctions made
political or human rights reform less, rather than more, likely.
19
By the early years of the
twenty-first century, the view that sanctions are successful only in relatively rare circumstances
was being described as the ‘traditional’ and ‘dominant’ view.
20
Those claims are sobering.
Let’s consider worries about the effectiveness of sanctions first, before turning to the question
of their proportionality.
17
Joy Gordon (1999) “A Peaceful, Silent, Deadly Remedy: The Ethics of Economic Sanctions,” Ethics
& International Affairs 13: 123-142.
18
John Mueller and Karl Mueller (1999) “Sanctions of Mass Destruction,” Foreign Affairs 78.3: 43-53,
at 43.
19
Peksen, Dursun and A. Cooper Drury (2010) “Coercive or Corrosive: The Negative Impact of
Economic Sanctions on Democracy,” International Interactions 36.3: 240-264; Oechslin, Manuel
(2014) “Targeting Autocrats: Economic Sanctions and Regime Change,” European Journal of Political
Economy 36.1: 24-40.
20
Jon Hovi, Robert Huseby, and Detlef Sprinz (2005) “When Do (Imposed) Economic Sanctions
Work?” World Politics 57.4: 479-499.

9
Effectiveness. International relations scholars may not agree about very much, but there does
appear to be consensus that sanctions fail far more often than they succeed.
21
The live
disagreement concerns the precise ratio of failure to success. On what is widely held to be the
most optimistic calculation, sanctions fail at least twice as often as they succeed.
22
An
influential reanalysis of the same cases suggested that the success rate was less than 5%,
however.
23
The authors later revised down their estimate, declaring that sanctions were partly
successful in effecting major policy change no more than 30% of the time.
24
In some of these
successful cases, note, the target countries were democracies. Indeed sanctions are sometimes
said to be more effective against democracies than against dictatorships.
25
One explanation is
that it is easier for outsiders to engender policy change in democracies, first because leadership
turnover is more routine (there are procedures for the regular rotation of leaders), and second
because the costs to defeated leaders are less severe. In dictatorships, by contrast, there are few
opportunities for leadership turnover, and turnover can have very serious consequences for
defeated leaders. Both facts make reform less likely in autocracies, other things being equal.
Moreover, it is widely believed that sanctions are often counter-productive. The
empirical literature sketches two key mechanisms here. First there is what we can call a
‘resource concentration’ effect. When dictators are starved of a stream of resources, they
21
Allen, Susan (2005) “The Determinants of Economic Sanctions Success and Failure,” International
Interactions 31.2: 117-138; Lektzian, David and Mark Souva (2007) “An Institutional Theory of
Sanctions Onset and Success,” Journal of Conflict Resolution 51.6: 848-871.
22
Gary Hufbauer, Jeffrey Schott and Kimberly Elliott (1985) Economic Sanctions Reconsidered, first
edition. Washington D.C.: Institute for International Economics, 93.
23
Robert Pape (1997) “Why Economic Sanctions Do Not Work,” International Security 22.2: 90-136.
24
Gary Hufbauer, Jeffrey Schott, Kimberly Elliott and Barbara Oegg (2007) Economic Sanctions
Reconsidered, third edition. Washington D.C.: Peterson Institute for International Economics, 159.
Note that these figures evaluate success in achieving a variety of ends, not all of which reduce to
improving democracy or human rights. There appears to be little literature assessing the effectiveness
of sanctions aimed at promoting democracy and human rights in particular, though Peksen suggests that
those aiming at promoting human rights are especially likely to be ineffective. Dursun Peksen (2009)
“Better or Worse? The Effect of Economic Sanctions on Human Rights,” Journal of Peace Research
46.1: 59-77.
25
Nikolay Marinov (2005) “Do Economic Sanctions Destabilize Country Leaders?” American Journal
of Political Science 49.3: 564-576.

10
concentrate what they do have even more exclusively on their immediate allies, in order to
secure their continued support.
26
Second, there is what we can call a ‘threat / repression’ effect.
Sanctions often encourage dissidents to mobilise; in many cases that is their point. But dictators
respond to such threats to their position by cracking down even more harshly on real or
perceived opposition.
27
Along either modality, sanctions can cause serious harms to the people
we are trying to help.
28
As a result, they have been shown to be associated with significantly
higher violations of physical integrity rights in the target country compared to the status quo.
29
Sanctions also frequently generate backlash effects, with leaders invoking nationalist rhetoric
to blame outsiders for the ensuing deprivation.
30
They appear to be associated with a
deterioration in the rule of law and an increase in corruption too, as citizens are driven to find
new sources of goods and / or income, with members of the elite often profiting from the new
opportunities this opens up.
31
Given the apparent consensus that sanctions only work in a minority of cases, it is not
surprising that considerable effort has been put into identifying what the most propitious
circumstances are in which to deploy them. Kaempfer, Lowenberg and Mertens have suggested
that sanctions are more likely to work ‘if there exists within the target country a reasonably
well-organized opposition group whose political effectiveness potentially could be enhanced
as a consequence of sanctions. In the absence of such a group, the sanctions might only
strengthen the regime’s pursuit of its objectionable policy by helping to rally public opinion
26
Gordon, “A Peaceful, Silent, Deadly Remedy,” 125.
27
Reed Wood (2008) “‘A Hand Upon the Throat of the Nation’: Economic Sanctions and State
Repression, 1976–2001,” International Studies Quarterly 52.3: 489-513.
28
Mueller and Mueller, “Sanctions of Mass Destruction.”
29
Wood, “A Hand Upon the Throat,” at 503; Peksen, “Better or Worse.”
30
Johan Galtung (1967) “On the Effects of International Economic Sanctions, With Examples from the
Case of Rhodesia,” World Politics 19.3: 378-416; Gordon, “A Peaceful, Silent, Deadly Remedy,” 134-
5.
31
Daniel Drezner (2011) “Sanctions Sometimes Smart: Targeted Sanctions in Theory and Practice,”
International Studies Review 13.1: 96-108; Wood, “A Hand Upon the Throat”; Lori Damrosch (1994)
“The Collective Enforcement of International Norms Through Economic Sanctions,” Ethics &
International Affairs 8: 59-75.

11
around the government.’
32
Escriba-Folch and Wright, meanwhile, suggest that sanctions make
‘personalist’ dictators somewhat more vulnerable to domestic opposition, but have no positive
effect (and perhaps a negative effect) in the case of single-party and military dictatorships.
33
Finally, it has been suggested that long-running dictatorships are the least plausible targets, at
least when it comes to political reform: ‘Enduring autocracies are least likely to justify the
imposition of broad and severe sanctions to promote democratic change,’ claim Marinov and
Nili. In such cases, ‘it is not clear how even a high cost imposed on the population at large can
translate into political action against the regime. The collective action problems inherent in
organizing against a despotic regime, when there are no recurring elections to serve as focal
points, are too formidable.’
34
In summary, the odds of success for economic sanctions do not
appear to be good, and negative outcomes are common. But effective sanctions packages may
include those targeting regimes which are relatively new, and / or personalist in character, and
/ or which already face major organised opposition.
Proportionality. Even if sanctions can sometimes achieve valuable goals, can they do so
without generating excessive moral costs? The challenge from proportionality appears morally
more grave than that of effectiveness. For a policy to fail to do good is one thing. But for it to
do more harm than good is more serious still. The emphasis on proportionality directs our
attention to comparative judgements about the effects of sanctions versus doing nothing on the
one hand, and versus other accessible courses of action on the other.
35
One implication of the
32
William Kaempfer, Anton D. Lowenberg, and William Mertens (2004) “International Economic
Sanctions Against a Dictator,” Economics & Politics 16.1: 29-51, at 31.
33
Abel Escribà-Folch and Joseph Wright (2010) “Dealing With Tyranny: International Sanctions and
the Survival of Authoritarian Rulers,” International Studies Quarterly 54.2: 335-359.
34
Nikolay Marinov and Shmuel Nili (2015) “Sanctions and Democracy,” International Interactions
41.4: 765-778, at 774.
35
James Pattison (2018) The Alternatives to War. Oxford: Oxford University Press, 6. In what follows
I will concentrate on the effects on the subjects of dictatorships. But if – as seems likely – sanctions
have costs for citizens of purchasing countries, those would have to be taken into account in any
proportionality calculation. So too would costs for third parties, which some sanctions scholars have

12
comparative nature of judgements about proportionality is that the fact that sanctions often
harm the innocent cannot by itself be considered a decisive objection against their use.
36
After
all, doing nothing also means that subjects will continue to experience abuses of their human
rights.
How, then, do sanctions fare in terms of proportionality? Generalised trade sanctions are
widely agreed to be a very blunt instrument of foreign policy. Their effects on domestic
populations are widespread and diffuse. If anything, they tend not to hurt the target regime, but
to hurt its opponents instead.
37
The resource concentration and threat / repression effects
provide the causal mechanisms here: dictators typically respond to sanctions by diverting still
more resources to their power base, and stepping up repression against dissidents. Sanctions
are often associated with a deteriorating human rights situation,
38
not least because they depress
the economy, worsen public health and education provision, and subject civil society to
increased repression.
39
On one well-known analysis, extensive sanctions make
‘disappearances’ of dissidents 115% more likely, incidents of torture 61% more likely, and
extra-judicial killings 64% more likely, with multilateral sanctions associated with a greater
negative effect in each case compared to unilateral sanctions. Worryingly, these accentuating
effects are more pronounced in the case of sanctions aimed at preventing human rights abuses,
when compared to sanctions aiming at other goals.
40
Sanctions have also tended to
disproportionately impact upon women.
41
According to Lori Damrosch, ‘The overwhelming
suggested can be significant. See e.g. Michael Canes (2000) “Country Impacts of Multilateral Oil
Sanctions,” Contemporary Economic Policy 18.2: 135-44.
36
Fabre, Economic Statecraft, 49.
37
Gordon, “A Peaceful, Silent, Deadly Remedy”: Drezner, “Sanctions Sometimes Smart.”
38
Yitan Li and A. Cooper Drury (2004) “Threatening Sanctions When Engagement Would Be More
Effective: Attaining Better Human Rights in China,” International Studies Perspectives 5.4: 378-394.
39
Thomas Weiss et al, “Economic Sanctions and their Humanitarian Impacts: An Overview,” in Weiss
et al (eds 1997) Political Gain and Civilian Pain: Humanitarian Impacts of Economic Sanctions.
Lanham: Rowman and Littlefield, 1-34.
40
Peksen, “Better or Worse?” 69-74.
41
Scott Wisor (2016) “Conditional Coercion versus Rights Diagnostics: Two Approaches to Human
Rights Protection,” Politics, Philosophy & Economics 15.4: 405-23, at 408; A. Cooper Drury and

13
impression remains that internationally sponsored sanctions have had the perverse effects of
enriching the targeted elites while simultaneously causing ever greater impoverishment of
civilian populations.’
42
Those we are trying to empower – members of the domestic opposition
– appear more likely to suffer than the average citizen.
None of this rules out the use of economic sanctions. If the alternative is to allow mass
atrocities or genocides to occur, then it has been argued that sanctions might be preferable even
if they are associated with an increase in serious human rights abuses - because sanctions might,
for instance, spread harms more thinly and evenly compared to mass killings.
43
But note, first,
that even if the comparative case can be made in instances where dictators are engaged in mass
atrocities and foreign aggression, it remains to be made in the case of what Michael Walzer
once called the ‘ordinary oppression’ typical of dictatorships
44
: the widespread deprivation of
political rights, the imprisonment of dissidents, and so on. These quotidian offences are our
topic in this paper, and here the comparative argument looks on its face to be harder to make.
Recall, second, that sanctions must be compared not only to doing nothing, but also to other
options, some of which might perform better on the score of proportionality. As such a
justification of sanctions must also involve an analysis of any other plausible and effective
responses to dictatorship – on which more later.
Improving the Performance of Sanctions?
The preceding section considered the prospects for cessation if what we care about are the
consequences for the oppressed subjects of dictatorships. We have seen that the likely
Dursun Peksen (2014) “Women and Economic Statecraft: The Negative Impact International Economic
Sanctions Visit on Women,” European Journal of International Relations 20.2: 463-90.
42
Damrosch, “The Collective Enforcement of International Norms,” 74.
43
Pattison, Alternatives to War, 66-7.
44
Michael Walzer (1980) “The Moral Standing of States: A Response to Four Critics,” Philosophy and
Public Affairs 9.3: 209-229, at 218.

14
effectiveness of the withdrawal of trade is highly uncertain, and that there are, meanwhile,
serious concerns about proportionality. There is room, of course, for disagreement about the
precise degree of responsibility we – as outsiders - bear for the negative outcomes which are
often associated with economic sanctions. Fabre suggests that even if we are merely
intervening agents in the negative humanitarian effects associated with sanctions, with the costs
being directly imposed by dictators themselves, then we are ‘still responsible in some sense for
those costs, and must take them into account when deciding whether sanctions are justified
overall.’
45
I believe Fabre is right here, but the crucial point for my purposes is the final clause
of that sentence. If we are judging sanctions on the basis of their instrumental effects, then
whoever is responsible for these harms, their scale needs to be taken into account in a reckoning
of the effectiveness and proportionality of sanctions.
Here, as we have seen, we have serious grounds for concern. It has been suggested that
these grounds are sufficient, in fact, to place pressure on the widespread idea that we ought to
see the deployment of military force as the last resort in cases of severe repression. In some
cases, military force might even be preferable to sanctions, if its effects are more concentrated,
shorter in duration, and / or more closely targeted at repressive elites rather than ordinary
citizens.
46
Whether that judgment is sound or not, these worries certainly justify a serious
investigation of alternative measures which might be associated with less damaging effects on
civilian populations.
i. Smarter Sanctions
Perhaps we can improve sanctions’ performance, however, by being more selective or ‘smart’
in their imposition. In his book Blood Oil, Leif Wenar argues that outsiders should break off
45
Fabre, Economic Statecraft, 54.
46
David Fisher (2011) Morality and War: Can War Be Just in the Twenty-first Century? Oxford: Oxford
University Press, 73.

15
purchases of oil (only) from societies where rulers are not even minimally accountable to
citizens in their decision-making about the resource trade. A Clean Hands Act would make it
illegal in domestic law for individuals or corporations to continue to purchase oil from
blacklisted countries. Should actors in other countries continue to do so, tariffs could be placed
on imports from those countries, with the proceeds being held in a Clean Hands Trust until
such time as the target country achieved a minimal degree of accountability in the resource
trade.
47
The proposal is therefore selective in two senses. First, the cessation of trade is
restricted to the ‘worst of the worst’ when it comes to accountability, targeting those regimes
which most clearly fail on publicly-available indexes of human rights or political inclusion.
48
Second, it focuses on specific goods which are closely associated with poor governance. The
primary focus is on oil, though in principle the proposal might be applied to other natural
resources associated with conflict or authoritarianism too.
49
It seems plausible on its face that in specifying the target of sanctions more closely in
these two senses, their performance on the criteria of effectiveness and proportionality might
be improved. In fact, however, serious worries remain in relation to both standards. When it
comes to effectiveness, we have already noted evidence that targeting the worst rulers is
unlikely to improve the ratio of success to failure. In fact the opposite appears to be the case.
50
47
Blood Oil, 289-91. Wenar in fact denies that his proposal amounts to an economic sanction: sanctions,
he claims, are punitive, whereas his proposal aims to realign incentives in the direction of greater
accountability. Leif Wenar (2018) “Reply to Blake and Mehdiyeva,” in Wenar et al, Beyond Blood Oil.
Lanham MA: Rowman and Littlefield, 119-31, at 124. In fact, sanctions have many goals aside from
punishment, and they very often aim precisely to discourage some form of behaviour (such as the abuse
of human rights, or external aggression), and to incentivise adherence to some norm or goal by holding
out the promise that restrictions on trade will be lifted once compliance is achieved. See Hufbauer et al
(2007) Economic Sanctions Reconsidered, 3-7; Fabre, Economic Statecraft, chapter 2; Pattison,
Alternatives to War, 40-2. As such evidence on the performance of sanctions appears relevant when
assessing these proposals.
48
Wenar, “Beyond Blood Oil,” in Wenar et al, Beyond Blood Oil. Lanham MA: Rowman and
Littlefield, 1-35, at 19.
49
Ibid, 20.
50
David Letzkian and Mark Souva (2007) “An Institutional Theory of Sanctions Onset and Success,”
Journal of Conflict Resolution 51.6: 848-71; Marinov, “Do Economic Sanctions Destabilize Country
Leaders?”; Marinov and Nili, “Sanctions and Democracy.”

16
One explanation is that democratic rulers fear being punished electorally if sanctions bite,
whereas dictators need not. A second explanation is that sanctions do little to alter the domestic
or regional balance of power which dictators inhabit.
51
As a result, authoritarian leaders are
said to have little incentive to concede to economic coercion in most situations we are likely to
confront.
52
It might well be easier to build international political support for the cessation of
trade with the worst regimes, but the evidence suggests that focusing on those regimes is likely
to mean a deterioration rather than an improvement in rates of success.
It might be thought, however, that focusing sanctions on one good, such as oil, will pay
dividends. If oil sales fund military expenditure, for instance, we might suppose that sanctions
will serve to reduce that expenditure and hence leave dictators more vulnerable to reform
pressure. In fact, the opposite appears to be the case. Sanctioned dictators increase military
spending – probably in response to perceived threats - and they do so by reducing spending on
public goods for ordinary citizens: ‘when survival is at stake, developmental concerns are left
aside.’
53
Sanctions on oil-producing countries appear to be even less likely than average to
destabilize rulers.
54
Van de Graaf has suggested that oil sanctions are most plausible in specific
economic conditions where buyers exert greater-than-usual power, and where the oil price
already happens to be unusually low. In other cases, he suggests, sanctioned countries can
usually ride out oil sanctions without too much difficulty.
55
51
Wisor, “Conditional Coercion versus Rights Diagnostics,” 408.
52
Susan Allen (2005) “The Determinants of Economic Sanctions’ Success and Failure,” International
Interactions 31.2: 117-138, at 133.
53
Abel Escribà-Folch (2012) “Authoritarian Responses to Foreign Pressure: Spending, Repression, and
Sanctions,” Comparative Political Studies 45.6: 683-713, at 699. Hultman and Peksen also report that,
whereas arms embargoes may reduce violence within target countries, sanctions in general appear to
correlate with increased violence. Lisa Hultman and Dursun Peksen (2017) “Successful or
Counterproductive Coercion? The Effect of International Sanctions on Conflict Intensity,” Journal of
Conflict Resolution 61.6: 1315-39.
54
Escribà-Folch and Wright, “Dealing With Tyranny,” 355.
55
Thijs Van de Graaf (2013) “The ‘Oil Weapon’ Revisited: Sanctions Against Iran and US-EU
Structural Power,” Middle East Policy 20.3: 145-63, at 146.

17
There does, however, appear to be at least once case in which oil sanctions have been
successful in shifting state policy. Wenar notes that oil sanctions on Iran appear to have been
pivotal in engendering a shift towards denuclearisation.
56
But it is useful, I suggest, to look at
the Iranian case in a slightly broader context. Sanctions on Iranian oil exports have now been
in place for four decades. The US has not bought a single barrel of Iranian oil since 1991, and
has placed considerable pressure on third parties to conform to that policy.
57
It does appear that
the precipitous drop in global oil prices in 2014, which seriously weakened Iran’s economy,
encouraged it to seek an easing of sanctions.
58
Decades of sanctions have signally failed,
however, in their principal and longstanding objectives: to deter Iran from sponsoring
terrorism, and to push it down the road to political reform.
59
At the time of writing, the Iran
nuclear deal is under considerable pressure, and Iran has reportedly stepped back from key
nuclear-related commitments.
60
But whether it ultimately succeeds or fails, we are still waiting
for examples of oil sanctions engendering political inclusion or human rights reform.
We must also address concerns about proportionality; but here too oil sanctions
experience difficulties. When Fabre discusses the cessation of sales of goods to brutal regimes,
she suggests that the case is going to be easier to make in the case of ‘single-purpose goods’
which are key to regime survival – such as weapons – and all but impossible to make in the
case of ‘dual purpose goods’ which can be used to advance repressive and less objectionable
56
Wenar, “Reply to Blake and Mehdiyeva,” 124.
57
Van de Graaf, “The ‘Oil Weapon’ Revisited,” 147.
58
There also appears to have been a greater willingness, especially in Europe, to strike a deal in part
because of the ongoing humanitarian impact of sanctions. Critics within the European Union suggested
that oil sanctions were having such broad and indiscriminate effects that, though technically ‘targeted’
or ‘smart,’ these sanctions in practice reproduced the effects of comprehensive sanctions - which the
EU had repudiated as a legitimate foreign policy instrument. Moret, “Humanitarian Impacts of
Economic Sanctions,” 120.
59
Jeffrey Schott (2012) “Economic Sanctions Against Iran: Is the Third Decade a Charm?” Business
Economics 47.3: 190-192, at 191.
60
This has led to suggestions that sanctions will resume once again. Jean-Baptiste Vey and Ahmed
Rasheed, “France Suggests Sanctions Could be Reimposed if Iran Reneges on Deal,” Reuters 7th May
2019.

18
goals. As a result of uncertainty about the humanitarian effects of generalised sanctions,
‘sanctioning parties must opt for highly targeted sanctions – with respect to their objects and
the individuals against whom they apply.’
61
But when we buy goods from dictators, we send in
the other direction the dual purpose good par excellence: money. Cutting off a major source of
foreign income therefore appears highly likely to impact negatively on the civilian population.
The ‘resource concentration effect’ suggests that the cessation of purchases will diminish the
provision of public goods.
62
In practice, oil sanctions do appear to be associated with significant
economic downturns and large-scale reductions of state spending on civilians.
This effect appears to have been borne out in every major recent case of oil sanctions. In
Iran, oil sanctions led to medical shortages which are widely believed to have caused many
additional deaths. Those suffering from cancer experienced severe downturns in their access to
lifesaving treatments,
63
as have those suffering from haemophilia, thalassemia, and other
complex diseases.
64
Restrictions on oil sales from Iraq in the 1990s were associated with many
tens of thousands of civilian deaths.
65
As spending on health and sanitation declined, sanctions
coincided with steep rises in tuberculosis, measles and typhus.
66
They also generated
61
Fabre, Economic Statecraft, 51.
62
Escribà-Folch, “Authoritarian Responses to Foreign Pressure,” 705; Damrosch, “The Collective
Enforcement of International Norms,” 74; Weiss et al, “Economic Sanctions and their Humanitarian
Impacts”; Hultman and Peksen, “Successful or Counterproductive Coercion?”; Dursun Peksen (2011)
“Economic Sanctions and Human Security: The Public Health Effect of Economic Sanctions,” Foreign
Policy Analysis 7.3: 237-51.
63
Shohreh Sahabi et al (2015) “The Impact of International Economic Sanctions on Iranian Cancer
Healthcare,” Health Policy 119: 1309-18, at 1310.
64
Mehran Karimi and Sezaneh Haghpanah (2015) “The Effects of Economic Sanctions on Disease
Specific Clinical Outcomes of Patients with Thalassemia and Hemophilia in Iran,” Health Policy 119:
239-43.
65
Michael Blake cites the famous claim that oil sanctions led to half a million additional child deaths.
Blake, “Bad Men and Dirty Trade,” in Wenar et al, Beyond Blood Oil, 37-49, at 41. That figure has
been seriously undermined by subsequent empirical work. But even critics of the figure accept that
child mortality increased significantly during the sanctions period. See Tim Dyson and Valeria Cetorelli
(2017) “Changing Views on Child Mortality and Economic Sanctions in Iraq: A History of Lies,
Damned Lies and Statistics,” BMJ Global Health 2.2: 1-5.
66
Karen Morin and Stephen Miles (2000) “The Health Effects of Economic Sanctions and Embargoes,”
Annals of Internal Medicine 18.132: 158-61. Because of their very broad humanitarian consequences,
Fabre considers Iraq the ‘paradigmatic example of unjustified sanctions.’ Fabre, Economic Statecraft,
75.

19
significant economic declines for third party countries dependent on oil imports or whose
economies were closely tied to Iraq’s.
67
The oil embargo in Syria also appears to have
contributed to a severe economic downturn, increasing unemployment, and triggering
significant cuts to welfare benefits. Water supply and sanitation have suffered, leading to an
increase in deadly waterborne infections, especially in children,
68
and people suffering from
long-term health conditions have again experienced great difficulties in accessing life-saving
medicines.
69
Meanwhile figures close to the elite have benefited considerably from new black
market opportunities. Sanctions do not, however, appear to have driven the Syrian regime to
engage in political reform.
70
The same can be said for the cases of both Iraq and Iran.
I do not claim here that the emerging evidence is decisive. In fact, when assessing the
impact of breaking off oil purchases we must confront a lack of systematic evidence. Oil
sanctions have not often been employed, and remain rather ill-studied.
71
But it is clear that in
every major case since the end of the Cold War, oil sanctions have led to significant numbers
of civilian deaths, including deaths of many children and of others unconnected to the ruling
regime.
72
Meanwhile, oil sanctions have not yet led to any significant instances of political
inclusion or human rights reform.
On the basis of both effectiveness and proportionality, therefore, the proposal that we
ought to break off oil purchases experiences difficulties. Whereas success is highly uncertain,
67
Canes, “Country Impacts of Multilateral Oil Sanctions,” 142-3.
68
Waleed al Faisal (2012) “Syria: Public Health Achievements and Sanctions,” The Lancet 379: 2241.
69
Kasturi Sen et al (2012) “Syria: Effects of Sanctions and Conflict on Public Health,” Journal of Public
Health 35.2: 195-9.
70
Peter Harling (2012) “Collectively Failing Syrian Society,” Foreign Policy 24th January.
71
Van de Graaf, “The ‘Oil Weapon’ Revisited,” 145.
72
Wenar’s response to arguments of that type is to raise the question of moral baselines. The status quo
in world politics – according to which we may buy oil from unaccountable leaders - also, he points out,
leads to many civilian deaths. Wenar, “Reply to Blake and Mehdiyeva,” 122. Note however that the
deaths which are said to have been caused by sanctions on oil-exporting countries like Iraq are deaths
over and above the level of deaths statistically ‘expected’ in the brutal regimes in question. See e.g.
Dyson and Cetorelli, “Changing Views on Child Mortality and Economic Sanctions.”

20
significant harms to civilians appear to be very likely. It might, to be sure, be the case that any
humanitarian suffering can be justified on the basis of long-term progress.
73
Perhaps the march
to greater accountability is simply a very long one which will eventually pay dividends. I do
not seek to rule out that possibility. But civilian deaths cannot be outweighed by hope alone:
any argument along these lines must show that the chance of significant progress for many
people is capable of outweighing the very likely harms to the people who are likely to die, or
experience severe suffering, as a result of oil sanctions.
74
Here there are two distinct challenges.
First, we must show that the progress in question is capable of morally outweighing the likely
harms. Defenders of so-called ‘Limited Aggregation’ have argued that grievous harms to small
numbers of people can only be outweighed by avoided harms to larger numbers of people if
the latter harms are, individually, at least similarly grave. The defender of sanctions must
engage with this challenge, by showing that gains in many peoples’ political inclusion can
morally outweigh even small numbers of additional deaths. Second, there is the epistemic
challenge. At present, we have several cases in which a policy of oil sanctions has intensified
suffering, but we do not yet have any cases where the result has been a transition to more
inclusive or accountable governance. Whereas humanitarian costs appear all but certain, in the
absence of successful real-world cases we do not yet know how to assign reliable probabilities
to the prospects of lasting political reform. This makes the weighing of expected harms and
expected benefits difficult to say the least.
75
ii. Threat versus Imposition
73
Wenar, Blood Oil, 298-9.
74
See for instance T. M. Scanlon (1998) What We Owe to Each Other (Cambridge, Mass.: Harvard
University Press), 239–40; Frances Kamm (2007) Intricate Ethics (Oxford: Oxford University Press),
484–86; Alex Voorhoeve (2014) “How Should We Aggregate Competing Claims?” Ethics 125: 64–87;
Larry Temkin (2012) Rethinking the Good (Oxford: Oxford University Press), 33-37.
75
For a sophisticated account of how to make proportionality calculations in cases where likely harms
and goods are uncertain, see Patrick Tomlin (2019) “Subjective Proportionality,” Ethics 129: 254–83.

21
There is a second way of responding to concerns about the effectiveness and proportionality of
sanctions. This is to highlight the potential systemic effects of threats. It has been suggested
that the threat of sanctions can be almost as effective in securing policy change in target
countries as their actual imposition.
76
Regimes which are vulnerable to the application of
economic pressure may choose to comply as soon as sanctions are threatened; regimes which
fail to respond appropriately to threats, by contrast, are scarcely more likely to respond
positively to their actual imposition.
77
But crucially, threats of sanctions, as opposed to their
imposition, may be associated with far fewer negative humanitarian effects in target countries.
We would not, of course, expect threats to be effective if they are never followed through on.
But if the threatened sanctions are eventually imposed in some cases of non-compliance with
human rights norms, there might be an overall deterrent effect when it comes to human rights
abuses.
78
Threatening to break off trade with dictatorships might even remove an incentive for
such regimes to emerge in the first place.
79
For this argument to succeed as a defence of the imposition of sanctions in those cases
requires us to weigh the negative effects on civilians of applying sanctions sometimes – a
76
Dean Lacy and Emerson Niou (2004) “A Theory of Economic Sanctions: The Role of Preferences,
Information and Threats,” Journal of Politics 66.1: 25-42; Drezner, “Sanctions Sometimes Smart”;
Hovi, Huseby, and Sprinz, “When Do (Imposed) Economic Sanctions Work?”
77
Hovi, Huseby, and Sprinz, “When Do (Imposed) Economic Sanctions Work?”, 379.
78
Though I cannot develop it here, there is an interesting parallel in this respect between the case of
sanctions and the case of the prosecution of dictators once they leave office. Threatening prosecution
may make dictators more likely to engage in violent repression in the short and medium term. But a
regime in which dictators are often prosecuted may make the emergence of dictatorship – or the
commission of the most serious abuses - systemically less likely. For a discussion of this issue see
Kathryn Sikkink (2011) The Justice Cascade: How Human Rights Prosecutions Are Changing World
Politics (London: W.W. Norton).
79
Pogge, World Poverty and Human Rights, 158-61. In fact, many empirical scholars are less certain
than Pogge that our willingness to buy resources does serve to incentivize coup attempts. Michael Ross
suggests that the chances of success are so low, and the up-front costs so high, that it is implausible that
coup attempts are motivated by a ‘honeypot’ effect. Ross, The Oil Curse, 161. For further claims that
oil wealth does not correlate with the probability of coups, see Anca Cotet and Kevin Tsui (2013) “Oil
and Conflict: What Does the Cross Country Evidence Really Show?” American Economic Journal:
Macroeconomics 5.1: 49-80, and Paul Collier and Anke Hoeffler (2005) “Coup Traps: Why Does
Africa Have So Many Coups D’etat?” Department of Economics Working Paper, University of Oxford.

22
necessary part of the deterrent effect – against the positive effects on civilian populations of a
broader regime of credible threats of sanctions. A regime of threats accompanied by selective
imposition might well be found, on the score of proportionality, to be more defensible than a
thoroughgoing policy of imposition.
Can this response serve as a defence of the across-the-board cessation of trade with
dictators? It seems clear that it cannot, and indeed is not designed to. The response makes
traction on the score of proportionality precisely by conceding that in many or most cases
sanctions will not be applied. We might generate a generalised movement towards human
rights observance, so the response suggests, by sanctioning some dictators and merely
threatening the rest. While such a policy cannot therefore justify a general policy of cessation,
it might still suffice to justify instances of cessation. Any such justification must, however,
explicitly identify who is likely to suffer and who is likely to benefit from the policy. It is
widely thought to be easier to justify harms to the innocent if the same innocent actors are the
intended beneficiaries of a policy (such as the imposition of sanctions). But in the case of a
policy of deterrence, the beneficiaries are likely to be distinct from those who will suffer harms.
The beneficiaries, it seems, will be individuals in other states. Scholars who have pinpointed
the efficacy of threats argue that it is precisely the threat of sanctions, rather than their
imposition, which is effective in changing behaviour in the countries which are at the receiving
end of those threats. Imposition is a necessary backdrop to a policy of issuing realistic threats,
but there is no good reason, defenders claim, to expect that the eventual imposition of sanctions
will have any positive effects in the countries to which they are applied. To the contrary, these
scholars suggest that if credible threats have already failed, it is unlikely that imposition will
succeed in changing the direction of policy.
80
The beneficiaries of a policy of threats backed
80
Lacy and Niou, “A Theory of Economic Sanctions,” 27; Hovi, Huseby and Sprinz, “When Do
(Imposed) Economic Sanctions Work?”

23
by selective imposition, then, will be the citizens of other countries who may see an
improvement in their human rights situation. By contrast, it will be the citizens of countries
where sanctions are imposed who will in all likelihood bear the costs of making that regime of
threats credible.
Such a policy will find very little support from non-consequentialist theories of just war.
It might be thought that wars aimed at punishing some states for wrongdoing are capable of
being justified on wider deterrent grounds. But in fact, it appears that wars are formidably hard
to defend from such a perspective, precisely because they harm innocent civilians who are not
appropriate objects of punishment.
81
If the cautionary notes we have thus far sounded about
the humanitarian effects of sanctions are sound, then the selective imposition of sanctions on
deterrent grounds appears to face the same problem. But even on purely instrumental grounds,
the policy of deterrence may be hard to justify, simply because it involves predictable reversals
to the position of some of the most oppressed people in the world. To sum up, then, a
recognition of the deterrent role of sanctions is not capable of justifying the imposition of
sanctions on all dictators. If it is to justify imposition in some cases, this justification would
have to lean on the wider systemic effects of a regime of credible threats, rather than any
evidence that sanctions will succeed - where threat has failed – in helping the citizens of the
countries they are imposed upon. But here it seems that defending the likely harms to innocent
civilians on the basis that they might improve the prospects of outsiders is not an easy task.
Non-Instrumental Arguments
81
Victor Tadros (2014) “Punitive War,” in Helen Frowe and Gerald Lang (eds) How We Fight. Oxford:
Oxford University Press, pp. 18-37, at 32-34.

24
Given the apparently limited prospects that imposing economic sanctions will succeed in better
protecting dictators’ subjects from abuses, empirical scholars have expressed puzzlement that
states resort to them at all – and, indeed, resort to them often.
82
One possible explanation is that
outsiders feel they ought to cease trade because they are obliged to send out a signal to dictators
– and to others who might engage with them – that they are in some sense moral outlaws.
‘Economic sanctions are not used merely to block or deter the commission of human-rights
violations,’ as Fabre notes. ‘They are also used as a way to express condemnation of those
deeds.’ The thought is that ‘Faced with such grievous wrongdoing, we must at least do
something.’
83
This suggests an expressive argument for sanctions.
Nili, too, is clear that our duty to break off trade with dictatorships holds ‘independently
of any prospects of improved outcomes for others.’ And this is fortunate, on his view, because
advocates of sanctions are unable to show with any confidence that they will make things better
for the oppressed.
84
The reason for cessation on his account, though, is the preservation of the
moral integrity of liberal democratic societies. We believe that citizens ought to be participants
in their own self-government – including decision-making about the sale of precious resources
– and as a result trading with countries where governments do not involve their own subjects
in the relevant decisions is an exercise in hypocrisy. Such a reason for cessation stands even if
we have no grounds for confidence that the poor will be made better off, or given greater access
to the reins of political power. ‘Our own decency,’ as Nili puts it, ‘must be the grounds’ for
cessation.
85
A third non-instrumental basis for cessation is also possible. Below I will consider a
consent- or property-based argument, which suggests that we should cease our purchase of
82
See e.g. Pape, “Why Economic Sanctions Do Not Work.”
83
Fabre, Economic Statecraft, 62.
84
Nili, “Conceptualizing the Curse,” 112.
85
Ibid, 105.

25
certain goods – here, forms of public property – simply because they are not the property of
dictators. Respect for basic liberal principles such as consent renders our purchase of goods
from anyone other than their owners illicit.
My aim in this section is not to adjudicate between expressive, integrity-based or consent-
based justifications, but to assess the normative weight that any such argument will possess.
Whereas standards of effectiveness and proportionality appear clearly relevant to instrumental
arguments, it is not immediately obvious that the same can be said for non-instrumental
arguments. When it comes to expressive arguments, in particular, the criterion of effectiveness
struggles to find a foothold. Consider the cause of insurrection against a despot. We might
justify otherwise futile acts of resistance against tyrants on the grounds that they will one day
serve to embolden rebels with more realistic prospects of success, for instance. Revolutionary
efforts with little chance of coming to fruition have been defended on this basis.
86
But if the
distinction between instrumental and non-instrumental justifications is to remain clean, the
justification for cessation cannot be that it will serve longer-term goals. Cessation will be
justified even if it does not make political change more likely, whether now or later. Construed
as an expressive act at least, it is difficult to know what it would even mean to apply standards
of effectiveness to such an act,
87
beyond the basic question of whether the relevant idea is
clearly communicated.
But the standard of proportionality is still, I would argue, highly relevant. Sticking to our
moral principles may sometimes be valuable even if it advances no-one’s wellbeing. But it
cannot be justified regardless of the scale of costs to human wellbeing. In this section my focus
will be on the costs that non-instrumental justifications appear capable of justifying. This
question loses interest if, as Nili has suggested, things really cannot get any worse for the
86
Lea Ypi (2014) “On Revolution in Kant and Marx,” Political Theory 42.3: 262-287.
87
Cf. Fabre, Economic Statecraft, 63.

26
inhabitants of at least some dictatorships.
88
But in the last section we noted evidence that things
often can, in fact, get worse for many of the victims of dictators. Even in a brutal regime which
wholly excludes its subjects from power, significant setbacks to those who are excluded should
still be treated as morally very weighty. If it is true that torture, disappearances, and extra-
judicial killing spike in response to the imposition of sanctions,
89
then we need some way to
assess whether those costs are justified.
90
Weighing negative outcomes against non-instrumental gains is certainly not easy. But
that does not obviate the need to try. Joy Gordon has argued that using sanctions to send a
moral message (such as the condemnation of tyrants) is not justifiable if it comes at the cost of
hurting vulnerable people.
91
Given the findings we have discussed in this paper, this would
appear to render expressive arguments for sanctions largely toothless. Though initially
attractive, Gordon’s view is a little too simple however. There must be cases where small costs
to victims are a reasonable price to pay for repudiating the practices that harm them. But there
must also come a point where the costs incurred become too high. One implication of Nili’s
view is that we can justifiably pursue cessation even if it makes conditions somewhat worse
for the subjects of dictatorships. But where can we locate the limit case? On Nili’s view the
limit arises once cessation hinders physical survival on a massive scale.
92
Here I would say two
things. First, there are grounds for believing that any across-the-board cessation of trade will
88
Shmuel Nili (2019) “Global Poverty, Global Sacrifices, and Natural Resource Reforms,”
International Theory 11.1: 48-80, at 73.
89
Peksen, “Better or Worse?” 69-74.
90
Another possibility is that we might implement side-policies to offset any negative effects. Nili
endorses my suggestion that if we were to implement Wenar’s proposals, we would also need to
implement side-policies mitigating any negative humanitarian consequences (Nili, “Global Poverty,
Global Sacrifices,” 73. See Armstrong (2017) Justice and Natural Resources: An Egalitarian Theory.
Oxford: Oxford University Press, chapter 7, section 2). Nevertheless, it is questionable whether the
‘targeted injections of funds and goods’ Nili has in mind will suffice to offset the costs incurred by
people who have been tortured or disappeared. If not, then we must still engage with the question
pursued in this section, of how much hardship non-instrumental arguments can justify.
91
Gordon, “A Peaceful, Silent, Deadly Remedy,” 138.
92
Nili, Conceptualizing the Curse,” 115.

27
cost many lives. If, as Nili suggests, this cost is too high, then cessation will be ruled out. But
second, the suggested standard seems to set the moral bar too high. It is not obvious why our
integrity is weighty enough to justify cessation even if the effect is merely to significantly
increase arbitrary imprisonment and torture, or to worsen diet and health outcomes across the
board, say – or indeed to make political reform less likely.
I believe that we can reach a similar judgement when it comes to consent- or property-
based justifications. On such views, outsiders should break off trade not because doing so will
promote democracy or human rights more generally, but because proper respect for the
property rights of people living in dictatorships demands it. If ‘the people’ own domestic
natural resources, then for outsiders to buy them from dictatorships without the people’s
consent appears to violate their property rights.
93
This standard for triggering cessation does not concern, to be clear, whether citizens
benefit from resource sales, but whether they have appropriately authorised those sales (in
principle they might consent but not benefit, or indeed benefit without having consented). Here
the details of the relevant proposals at hand may make a difference to how we assess the
normative weight of consent. Nobody, to my knowledge, has suggested that the actual consent
of the people to specific sales of public property – such as natural resources – is necessary for
trade to be permissible. If it were, international trade would presumably grind to a halt –
because not even in democracies are citizens consulted as a matter of course about such sales.
For Pogge, we are obliged to cease buying public property from dictators if the people have
93
Wenar, “Realistic Reform of International Trade and Resources,” in Alison Jaggar (ed) Thomas
Pogge and His Critics (Cambridge: Polity), 123-150, at 137. Commentators have sometimes wondered
whether Wenar’s defence of public property is merely pragmatic in character. See for instance Anna
Stilz (2018) “Does a Country Belong to Its People?” in Leif Wenar et al, Beyond Blood Oil (Lanham
MA: Rowman and Littlefield), 89-103, at 91. Shmuel Nili suggests that Wenar still owes us a deep
moral defence of public property of the type Nili himself provides. See Nili (2019) “The Idea of Public
Property,” Ethics 129: 344-369, at 346. I do not seek to resolve this issue here, because whether the
defence of public property is deep or merely pragmatic, we still need to know what kinds of costs can
justly be incurred in deference to it.

28
previously, and democratically, passed a constitutional clause explicitly indicating that the
emergence of dictatorship would vitiate their consent. An independent Democracy Panel could
then judge difficult cases.
94
For Wenar, cessation is mandatory if citizens have not had the
opportunity to give their opinion one way or another, in suitably un-coerced circumstances –
involving the free exchange of information about sales, and the ability to object to them without
incurring severe costs.
95
Pogge’s proposal does not yet generate any immediate action-guiding conclusions for
outsiders, because no country in the world has passed the kind of constitutional clause he has
in mind. His argument is primarily a suggestion that they ought to do so.
96
But we can still
inquire what normative weight such a reform would have should it come to pass in the future.
The kind of explicit statement Pogge imagines does appear to give us a powerful reason for
ceasing purchases should a dictatorship subsequently come to power. Even here, though, I am
not sure it would always provide a decisive reason, all-things-considered. The decision whether
to cease trade should still, it seems to me, be reviewable both in light of our emerging
understanding of the effects of cessation on the oppressed, and in light of our understanding of
the performance of alternative strategies for encouraging the restoration of democracy – on
which more shortly. The decisions we make here cannot, it seems to me, be independent of the
humanitarian consequences. The kind of case Wenar has in mind, by contrast, does not involve
the explicit withdrawal of consent to trade. Rather it sketches a set of standards according to
which (explicit or tacit) consent can be judged to be operative. These are intended to be very
minimal standards – indeed Pogge suggests that they could be met in autocratic regimes
without too much difficulty,
97
in which case our duty to withdraw trade from dictatorships
94
Pogge, World Poverty and Human Rights, 161-2.
95
Wenar, Blood Oil, 228.
96
Pogge, World Poverty and Human Rights, 158-72.
97
Thomas Pogge (2010) “Responses to the Critics,” in Alison Jaggar (ed) Pogge and His Critics.
Cambridge: Polity, pp. 175-250, at 225.

29
becomes moot. Perhaps the best we can say is that absent such conditions, we do not know
whether the people consent or not. Nili conjectures that, in light of the likely consequences of
cessation, citizens of many resource-cursed dictatorships probably would consent to continued
trade if given the choice.
98
But we do not know. In general, though, the lack of explicit consent
one way or another seems to me to weigh less heavily than its explicit withdrawal. If so the
cautions I expressed about holding firm to cessation in Pogge’s case would apply a fortiori
here. Non-instrumental arguments do appear to give us reasons for taking cessation seriously,
and perhaps powerful reasons. But if the empirical findings we have considered thus far are
robust, non-instrumental arguments for sanctions on dictatorships will also be much harder to
make than their supporters commonly recognise.
Withdrawing or Extending Trade?
Thus far, I have argued that the case for cessation – that is, the imposition of a collective ban
on purchasing goods from dictatorships, whether this is to apply to all goods or only to some –
is more difficult to make than our critics have suggested. The discussion so far has, however,
focused on withdrawing purchases – stepping back, that is, once a trading relationship is
already in place. But it might be responded that my arguments do not show – and have not
attempted to show – that we do anything wrong merely by refusing to enter into such relations
in the first place. Where no trading relationship exists, my argument does not establish that we
have any moral reason to enter into one. In our globalised world, of course, pretty much all
states trade with each other in some way – even when some form of sanctions is in place. Hence
perhaps we should reformulate the response: even if we have reasons – however regrettable –
98
Shmuel Nili (2016) “Liberal Integrity and Foreign Entanglement,” American Political Science Review
110.1: 148-159, at 155.

30
to maintain existing trading relationships, perhaps we should refrain from extending or
deepening those relationships. I have not said anything yet to suggest that such a position would
be morally suspect.
I will not attempt to refute that possibility entirely. But we did note earlier that a full
appraisal of the case for sanctions requires us to also scrutinise other available options for
reforming dictatorships. All I want to note here, by way of closing, is that the empirical
literature on trade provides some grounds for believing that increasing trade may actually be
the most powerful weapon we possess by way of which to encourage the reform of
dictatorships. The overall relationship between democracy, human rights, and trade is a large
and highly contested topic, not least because some free market reforms imposed on developing
countries in the past have apparently been associated with worsening respect for human rights,
including physical integrity rights.
99
But ‘structural adjustment’ policies notwithstanding, it has
often been suggested that integration into global trade is associated with improvements in the
quality of governance,
100
and in the rule of law,
101
and that integration into global trade and
investment patterns is key to promoting respect for human rights,
102
including both physical
security rights and political participation rights.
103
Why might this be the case? At least two mechanisms have been suggested which are
plausible on their face. First, it has been suggested that gains from trade can help to smooth the
process of political reform by allowing rulers to offset any costs incurred by supporters as a
99
Rodwan Abouharb and David Cingranelli (2006) “The Human Rights Effects of World Bank
Structural Adjustment, 1981-2000,” International Studies Quarterly 50.2: 233-262.
100
Fahim Al-Marhubi (2005) “Openness and Governance: Evidence Across Countries,” Oxford
Development Studies 33.3-4: 453-471.
101
See e.g. Roberto Rigobon and Dani Rodrik (2005) “Rule of Law, Democracy, Openness, and
Income: Estimating the Interrelationships,” Economics of Transition 13.3: 533-564.
102
Clair Apodaca (2001) “Global Economic Patterns and Personal Integrity Rights After the Cold War,”
International Studies Quarterly 45.4: 587-602.
103
David Richards, Ronald Gelleny, and David Sacko (2001) “Money with a Mean Streak? Foreign
Economic Penetration and Government Respect for Human Rights in Developing Countries,”
International Studies Quarterly 45.2: 219-239.

31
result of reform, transforming reform from a zero sum into a positive sum outcome.
104
Martin
Ravallion has made a parallel case about the role of foreign aid in easing political transitions,
by way of which an injection of income can help overcome stumbling blocks on the road to
reform.
105
This claim, of course, might produce moral unease: we might well believe that
domestic elites should bear the costs of reform, and that compensating them for their pains is
far from being morally required. But granting that dictators and their allies are not entitled to
compensation is not the same thing as establishing that it is wrong to provide it, or indeed that
we have no duty to provide it. Rewarding evil-doers is a pro tanto moral bad. But it may be
preferable to allowing millions to continue to live under tyrants.
106
When it comes to the worst
abusers of human rights, the early steps towards reform may often involve some form of
instrumental bargaining, in the hope that this can sow the seeds of deeper change, including by
creating new constituencies which stand to benefit from further reform.
107
Second, trade – and especially trade in a diverse range of goods – facilitates the
emergence of a strong middle class.
108
As is familiar from many accounts of political
development, this emergence in turn generates claims for political inclusion.
109
Other things
being equal, trade in a more diverse set of goods looks like being more beneficial here than
trade in a narrow set. The more diverse the economy, the harder it is going to be for the ruling
regime to securely capture economic rents, the more sources of independent economic power
there are going to be, and the more powerfully demands for political inclusion are therefore
104
Emilie Hafner-Burton (2005) “Trading Human Rights: How Preferential Trade Agreements
Influence Government Repression,” International Organization 59.3: 593-629, at 607.
105
Martin Ravallion (2014) “On the Role of Aid in The Great Escape,” Review of Income and Wealth
60.4: 967-984.
106
See also Pattison, Alternatives to War, 150.
107
Thomas Risse and Kathryn Sikkink (1999) “The Socialization of International Human Rights Norms
Into Domestic Practices: Introduction,” in Risse et al (eds) The Power of Human Rights: International
Norms and Domestic Change. Cambridge: Cambridge University Press, 1-38, at 10.
108
Peksen, “Better or Worse?”
109
Seymour Lipset (1959) “Some Social Requisites of Democracy: Economic Development and
Political Legitimacy,” American Political Science Review 53.1: 69-105.

32
going to resonate. If it is indeed plausible, this story would provide further reason – beyond the
imperatives of avoiding dangerous climate change – for assisting or persuading oil-producing
countries along the road to economic diversification.
110
But rather than ceasing trade with oil
exporters, progress will likely involve fostering a wider variety of exports: enhancing and
extending, that is, rather than cutting, trading links.
I will not claim that the empirical evidence here is decisive and unequivocal by any
means. According to some commentators, for instance, even if openness to trade eventually
leads to benefits for citizens, the transition to greater openness can be fraught. Regimes which
are wholly closed to trade can be very stable, and greater openness can introduce a period of
greater instability before the long-term benefits of openness come to fruition.
111
There is at
least a plausible story, however, which suggests that extending and deepening trade can lead
to political reform. To take an even more fine-grained approach, one suggestion has been that
if we care about human rights, the most promising policy package would involve the provision
of benefits through Preferential Trade Agreements, with the realistic threat of the withdrawal
of those additional benefits (rather than the withdrawal of trade per se) hanging in the
background. This package emphasises carrots – and the removal of carrots – rather than sticks.
The most advanced example of this kind of policy has been the European Union’s Cotonou
Agreement, signed with 78 countries from Africa, the Caribbean and the Pacific, in which the
EU delivers non-reciprocal access to markets in some goods in return for specific human rights
commitments in vendor countries. Success in meeting these commitments is then regularly
reviewed, with preferential trading privileges withdrawn in serious cases of default. These
agreements ‘provide an economic motivation to promote human rights policy reforms that
110
For a discussion of the nature and basis of outsiders’ moral duty to promote post-oil development
options, see Chris Armstrong (2019) “Decarbonisation and World Poverty: A Just Transition for Fossil
Fuel Exporting Countries?” Political Studies (online early).
111
Ian Bremmer (2006) The J Curve: A New Way to Understand Why Nations Rise and Fall. New York:
Simon and Schuster, 3-26.

33
would not otherwise be implemented, and they do so in a relatively short time horizon.’
112
Cases where threats to human rights have been reversed in response to the combination of
threat and persuasion are said to include Togo and post-civil war Rwanda.
113
Preferential Trade
Agreements appear to have sustained democratization processes,
114
and Free Trade
Agreements more broadly appear to have been successful in numerous cases in preventing
unstable democracies from backsliding into autocracy.
115
One way to capture the different options open to us when dealing with dictatorships is to
distinguish ‘leverage’ – the use of diplomatic pressure, economic sanctions, and so on – from
‘linkage’ – the building up of greater and more diverse links, including trading links (but also
cultural, personal and intellectual links) between our own societies and those blighted by
dictatorship. According to Levitsky and Way, whereas leverage has typically failed, linkage
has often exerted an incremental but nevertheless significant influence in the direction of
democratisation. This, they claim, is for a variety of reasons: because with greater linkage,
abuses overseas become more visible within our own societies; because linkage helps to
inculcate democratic norms; because it builds the capacity of reform groups within
dictatorships; and because it gives those reform groups an international visibility which can
later protect them from abuse.
116
If this story is correct, then the conclusion for our purposes is
an interesting one. Perhaps, in practice, our best chance at encouraging the reform of
dictatorships is to trade with them more, rather than less. We might have reason not only to be
112
Hafner-Burton, “Trading Human Rights,” at 607.
113
Emilie Hafner-Burton (2013) Forced to be Good: Why Trade Agreements Boost Human Rights.
Ithaca: Cornell University Press, 143, 156.
114
Mark Manger and Mark Pickup (2016) “The Coevolution of Trade Agreement Networks and
Democracy,” Journal of Conflict Resolution 60.1: 164-191.
115
Xuepeng Liu and Emanuel Ornelas (2014) “Free Trade Agreements and the Consolidation of
Democracy,” American Economic Journal: Macroeconomics 6.2: 29-70.
116
Steven Levitsky and Lucan Way (2005) “International Linkage and Democratization,” Journal of
Democracy 16.3: 20-34.

34
circumspect about arguments for cessation, but to be open to extending trade links rather than
severing them.
The arguments of this section raise an important question about duties to engage in trade.
I have not claimed that there is a general duty to trade with other societies. In fact we have
reason to be cautious about any such argument, since the claim that there is such a duty has
been used to justify colonial wars and other abuses.
117
But it is highly plausible that we can
possess a derivative duty to trade with disadvantaged societies where that is useful in promoting
their development. If the advantaged possess a duty to ease poverty and suffering elsewhere,
118
an important question arises concerning the optimal means of acting upon that duty. One
familiar argument holds that we should prefer ‘trade’ to ‘aid,’ because trade better promotes
the autonomy of people in developing countries, rather than leaving them dependent upon the
often fickle and conditional delivery of aid.
119
If the empirical evidence canvassed in this
section is robust, it suggests that we have a second, and much less familiar, reason for preferring
trade as a policy option. If extending and deepening trade can help to create additional bases
of socio-economic power in dictatorships, as well as smoothing the path of political inclusion,
it may be a still more important vehicle of global reform than we have so far recognised.
Conclusions
There is no doubt that the question I have been tackling in this paper represents a major ethical
quandary, where any answer provokes considerable unease. The problem we face in
117
See for instance Anthony Anghie (1996) “Francisco de Vitoria and the Colonial Origins of
International Law,” Social and Legal Studies 5.3: 321-36.
118
In principle this duty might be based upon either capacity to help, contribution to the problem, or
both. For a careful analysis of these foundations for a duty to assist, see Christian Barry and Gerhard
Øverland (2016) Responding to Global Poverty: Harm, Responsibility, and Agency (Cambridge:
Cambridge University Press).
119
See for instance the discussion in Aaron James (2012) Fairness in Practice: A Social Contract for
the Global Economy (Oxford: Oxford University Press), chapter 7.

35
determining whether to continue, or terminate, trade with dictatorships in many ways resembles
that faced by humanitarian NGOs which must decide whether to continue to engage with those
dictatorships, or withdraw. Jennifer Rubenstein calls this the problem of ‘spattered hands,’ in
which actors must decide between two apparent injustices: ‘contributing to violence and
intimidation perpetrated by others, and “pulling the rug” out from under innocent civilians by
depriving them of basic services on which they have reasonably come to rely.’
120
My suggestion has been that claims that the answer here is straightforward – that
cessation is clearly morally required – are too quick. Insofar as we have evidence to go on from
the sanctions literature, the message is sobering. There is, it seems, depressingly little reason
to believe that disengagement will make things better for the oppressed, at least in the short
term. And it may well make things worse. We certainly ought to vigorously investigate any
critical levers we have which promise to make things better for the oppressed. But the signs
are that across-the-board disengagement is unlikely to be helpful. Depending on how the
evidence plays out, showing proper concern for the interests of the oppressed may well lead us
to the conclusion that we ought to deepen our trading relationships with dictatorships, rather
than rupturing them.
120
Rubenstein, Between Samaritans and States, 11.