A piece by Noam Scheiber in The New Republic , prompted me to get to work on a piece I’ve been meaning to write for ages, not so much because I have new and original ideas, but because I’d like to clarify my thoughts, with the help of discussion. The piece is subscription only, but the relevant quote is a point that’s been made before
The problem with [criticism of Bush’s handling of the Iraq war] is that there’s a difference between expecting the administration to fight a war competently and expecting it to fight an entirely different kind of war than the one you signed onto.
My starting point, then, is the observation that, in the leadup to the Iraq war there were numerous different cases for war, some publicly avowed at different times, and some not. These included WMDs, the War on Terror, humanitarian intervention, democracy promotion, the strategic importance of Iraq’s oil and simple vendetta. It might seem that the more reasons for war, the stronger the case, but the problem is that different cases for war imply different strategies for the war, and especially for the postwar period.
The ostensible basis for the war, WMDs, implied the need to act fast, since Saddam might use his weapons at any time, and implied a simple success condition: once the WMDs and the supporting infrastructure were found and destroyed, the US could withdraw and leave the Iraqis (minus Saddam) to sort own their own problems. Roughly speaking, this was the war we were sold, and this was the war we got, at least until it turned it there were no WMDs, and an early exit wasn’t really feasible.
Although the Iraq war seems to involve this problem to a high degree, it arises all the time. For example, there are a lot of different reasons for supporting reform of the House of Lords in the UK (the old structure was anachronistic, inefficient, anti-democratic, biased against Labour and so on), but they imply different kinds of reform.
Concern with democracy suggests an elected House, more representative than the Commons, where the first-past-the-post system turns minorities into majorities. By contrast, the main motive for the reforms around Blair seemed to be that hereditary legislators were bad for Britain’s image and occasionally obstructed Labour PMs – hence the preference for a weak upper chamber with appointed members.
I’m trying to think about a bunch of questions here.
* Are these problems more critical in relation to war than in other cases? I have a strong intuition that a decision to go to war should be based on a single sufficient casus belli, with any additional arguments being merely a bonus. So, as soon as it became clear, in late 2002 and early 2003, that the WMD case didn’t stand up, I opposed the war, taking the view that, if another case was to be made, the whole process had to be restarted. But I don’t have a clear view as to what it is that makes this requirement critical in relation to war.
* Is an analogy with criminal law useful? You can have a lot of evidence suggesting that X ought to be in jail, but we require proof beyond reasonable doubt of some particular offence.
* Do situations of this kind favour particular kinds of “lowest common denominator” solution, and if so how can we characterise them ?
* Has this question already been addressed? There’s not much in the parts of the decision theory literature with which I’m familiar, but that’s only a tiny subset.
Most importantly, of all
* When a decision is naturally posed in a binary form such as “War vs No War”, at what point do you decide that the multiple rationales problem is such as to undermine this, and draw a conclusion like “Not this war now”?
John Quiggin 
I opposed the war from the start and attended the anti war rally. This was not because I am/was anti Iraq war absolutely but for the reasons below. I knew before the war that:
A) There were no functioning WMDs and Iraq didn’t threaten anyone
B) The war would be conducted in a typically grossly incompetent fashion
C) All pre war reasons and justification from the leaders of the CotW were dirty filthy lies
D) My reasons and justifications for war were not theirs
E) The immense hypocrisy of it all, the US, Australia and the UK support many more regimes and have in the past that are even more brutal, undemocratic and corrupt than Hussien’s regime.
On that basis I take the position “Not this war now”. But war could have been just, fair and the right thing to do under different circumstances. In my view NATOs Kosovo bombing campaign was even more incompetent than the Iraq war.
War is more about power-politics than law-observance.
The state’s claim to go to war in a properly working democracy is usually framed in terms of benefits and costs in relation to the citizenry. In the case of an “improper” democracy, the case for war is more should be analysed in terms of the costs and benefits to state officials and their clients. In this case, the citizenry are simply a nuisance to the state.
In the US, the democratic process is fouled by crony-capitalist elitist-interests and Christian-fundamentalist populist-idealists. Hence Bush gets re-elected. Thus the US’s war in Iraq should be analysed for its cost/benefits to the Republican Party and its domestic and foreign clients.
In the case of the Iraq-war the ex-ante benefits of war seemed to be great and diverse at both national, regional and global levels.
The REPs, in the fraught post-911 era, could become the saviours of frightened Americans in the battle to control the US national state. A short, successful war against a perceived threat would enhance the national security credentials of the Daddy partyREPs against the Mummy party DEMs.
There were further regional power benefits to the civilian leadership PENTAGON. Wolfowitz theorised that knocking out Iraq – the major Baathist state in the ME which has been seen as a proliferator of WMDs, promoter of terrorism and frontline anti-ISRAEL – would allow it to be transformed into a pro-US (and proto-democratic) client state. The US admin thought this change necessary as the US’s traditional ME client state – Saudi Arabia – had gone on the nose after its role in 911 was discovered. Thus the US could, with one stone, deal a double blow by knocking out tyrannical Baathist Iraq and side-lining terroristic Wahabbist Arabia – which I call the ditch Saudis/hitch Iraqi strategy.
Finally there war the global power benefit seen in the US itself out-gunning its “peer competitors” at the UN (eg Old EU, PRC) and establishing itself as the global hegemon. The analogy with the US’s role in winning the Cold War, bringing down the USSR and emerging as a unipolar super-power are obvious.
These several benefits were all shared by various members of the US admin and its clients. But these people all shared one underlying belief: that the costs of tjhe war (fiscal, martial, diplomatic) would be minimal. The war would be short and self-funding (like GW I) and not-so bloody (like Afghan war). The US troops would be greeted as liberators. Chalabi would deftly transform Iraq into a secular pro-US client state withouth the need for prolonged occupation or reconstruction.
So when examining a possibly dodgy case for war, especially in an “improper” democracy, it is wise to ignore the putative pie-in-the-sky benefits and look at the down-to-earth cost schedule faced by the states “war-party”. If the state officials appear to be seriously underestimating/downplaying costs to the citizenry, or assume that the state’s costs will be minimal, then their war claims should be given thumbs down.
In the case of Iraq, the costs to the US Treasury and DoD are obviously enormous. And more so to Iraqi state and citizens. Watch the bottom-line!
This conservative principle goes double for any war of choice, rather than necessity.
Paul Krugman wrote a very interesting piece that is contained in The Great Unraveling about the changing rationale for the Bush tax cuts. When they were proposed, they were said to be because there was a surplus that needed to be given back to the people. When they were enacted, the rationale had changed to be that they were a stimulus package required to beat the recession. When they were lengthened, as they had originally been cut in weird ways to make them appear to cost less, they were, amazingly enough, put forth by Tom Delay as being important to the war effort.
Krugman points out that when something has a shifting rationale it is quite probable that there are real reasons that are not being told.
JQ, your comparison with democratic reform of the Lords is presumably intended to provide an established case where the behaviour and desired results are known by now. But it’s plain wrong – see my remarks at no. 7 of the latest weekend reflections.
It is not desirable to make two chambers do the same job. Rather, taking two different abstractions of the body politic is like using two different map projections; you know that neither will be accurate (see Arrow’s Theorem and some practical observation), but together they can be designed to mitigate the flaws the way plywood works as a composite material.
It is not an object of “good” government that it should be either efficient or effective, except in certain particular and well understood respects. To take it as a general truth is to confuse the two meanings of “good”, the ethical and the engineering ones (Eichmann was a good Jew killer: discuss). Many good – i.e. desirable – functions of a system are “inefficient” in the sense that they are deliberatley there to arrest or delay things; they are good like friction in brakes. Making the democratic process of a house of review “efficient” is as bad as oiling brake linings.
And “democracy” isn’t a good thing in itself anyway, but at best only instrumentally (and often not even that). It is certainly true that if you focus on the means and not the end you will eventually gravitate to a sort of attractor basin in which everything becomes self referring and the purpose of democracy is democracy, which loses the plot. Look around and you will see much evidence of that in the world today, and I don’t just mean failed democracies. The emergent behaviour of democracies is very different from “by the people, for the people, of the people”, and even that bgs a great many questions.
As a reminder, here are the three forms of incompleteness I have discerned in the theoretical framework of democracy:-
– It is inherently susceptible to selective editing and agenda control to achieve desired results, i.e. it is open to capture. It is not enough to dismiss instances of capture as examples of failed democracy, becuase on the one hand we have no mechanism available to handle that within the framework and on the other the theoretical framework itself always has an intrinsic vulnerability. It is the thing itself…
– It has no definition of “the people”, other than a circular one; it is entirely possible for democracies to elect a new people, though historically it has usually been gradual and, like all thin end of the wedge arguments, when it happens it is hard to notice precisely because it has succeeded.
– It cannot, of its nature, create or define values – ethics etc. – but at best express and at worst suppress them. Consider Thoreau’s dissenting “majority of one” over slavery, or the Athenians’ varied but instructive behaviour in Malos and Mitylene. (Hint: what they did in Malos led to its name changing to Melos.)
None of these meta-questions answer the questions of right or wrong either, of course, but they can be used to test supposed answers to these difficulties. For myself, even if I stipulate for the sake of argument that government is desirable, I still suppose that working answers for these areas must come from outside sources and only be implemented within democracy. Not to consider them leads to proceeding blindly with reforms that have neither practical, ethical, nor functional value. For those, look within yourselves, to the future, and to tradition, and edit out those things that do not work even by a self consistency standard. Back to Burke, say I.
And so much for democratic reform. It is enough that the system be democratic, not that any constituent part be (and that goes for you too, reflexive republicans).
“JQ, your comparison with democratic reform of the Lords is presumably intended to provide an established case where the behaviour and desired results are known by now.”
Au contraire, my whole point was that there were several different agendas for reform of the Lords, agreeing on the desirability of removing hereditary peers, but otherwise incompatible. The Australian Republic debate provides another instance.
Oscar Wilde pointed out that multiple excuses are always less convincing than one as they convey the impression that you’re making it up as you go along. The impression is correct in this case – they did make it up as they went along.
I do think you’re being entirely too kind to the Bushies – incompetence and mendacity are both reasonable explanations for their behaviour, and are not mutually exclusive. Those who believed our leaders and bought into the war ought now to be a bloody sight angrier at being gulled than those who, like me, opposed it from the start.
Not necessarily, derrida derider. Many of those who “bought into the war” did so for personal advantage, not because they believed the incredible, shifting set of excuses proposed by the US/UK Govts. Many have profited, and have no reason to be “a bloody sight angrier”.
And, Jack Strocchi, you forgot what may be a very important benefit to the military. Without oil, the tanks don’t move, the planes don’t fly and the ships don’t leave harbour. The modern armed forces become museum pieces and nobody is interested in what the generals say anymore. I would imagine the Pentagon is very worried indeed about somebody working this out and suggesting a return to cavalry. How many generals and military contractors know anything about horses?
My responses to your 4 questions, JQ:
1. Are these problems more critical in relation to war than in other cases?
The answer depends on the values of the person answering the question. I doubt any society could reach a total consensus on what the most important issues are, even at a time of War.
2. Is an analogy with criminal law useful?
One of the main tenets of contemporary argumentation theory is that valid and appropriate forms of argument are domain-specific. (This contrasts with formal logic and contemporary decision theory, which claim to have developed domain-independent methods of reasoning. )
So I would respond to this question with: Analogy with any other domain may be useful, but should not be seen as determining the answer in the original domain (here, public policy decision-making). Other domains of relevance may be: medicine, civic planning, and complex engineering design (arguments for/against design options), all areas of study by argumentation theorists.
3. Do situations of this kind favour particular kinds of “lowest common denominator� solution, and if so how can we characterise them ?
I don’t see why this should be so. After all, one of the achievements of the Hawke and Keating Labor Governments was to put quite complex economic proposals o the public and make convincing cases for them. I think Blair, Bush and Howard distrusted public opinion too much to actually try to convince us by reasoned argument.
4. Has this question already been addressed?
Yes, but not in the decision theory literature, which has been dominated for the last half-century by narrow-minded and historically-unaware economists and statisticians. The relevant literature is that of argumentation theory, a branch of philosophy begun by Aristotle, revived in the 1950s by Toulmin, Perelman and Hamblin, and now an important subject in computer science. (The rise of the Internet has led computer scientists to focus attention on intelligent computational entities reasoning with one another in order to achieve their individual goals.)
The question you are asking is understood under the names of “argument aggregation” and “argument accrual”, and has received some recent attention, but usually in specific domains. Henry Prakken, a computer scientist at Utrecht University, last month presented a paper at the International Conference on AI and Law (ICAIL) in Bologna, on the topic of argument aggregation in legal domains. He proposes three principles of accrual of arguments. I am not sure I agree with these, but they may be helpful to you in thinking about the question of accrual in the public policy domain.
JQ, isn’t that precisely what I meant about reflexive?
I forgot Q5:
5. When a decision is naturally posed in a binary form such as “War vs No War�, at what point do you decide that the multiple rationales problem is such as to undermine this, and draw a conclusion like “Not this war now�?
I think that most academics who study argumentation would say that the more arguments there are in favour of a claim, the greater are the reasons for supporting the claim. Indeed, this principle is the basis of most of the various formalisations of argumentation used in computer science. So, on this basis, my answer to your specific question would be “Never”.
I think what you are doing (implicitly in your question 5) is mixing argument about the arguments for/against going to war, with argument about the motivations of the people making the arguments for/against going to war. Your reason above for rejecting the case for war (it seems to me) is not really because there are multiple arguments for doing so, but because the presence of multiple arguments in favour leads you to distrust those making the arguments in favour. Your conclusion may still be correct (and justifiable and rational, etc), but you should be clear that this is what you doing — arguing about the motivations of your opponents, not about the arguments in favour or against the proposal.
For myself, I see no rational reason to reject a proposal for action simply because a multitude of arguments in its favour have been made by its proponents. Some of these arguments in favour may have been made in response to attacks on earlier arguments in favour. That a proponent is willing to engage in debate and to modify an earlier justification in response to attacks received would be seen by most people as a desirable feature of a democracy, not a negative.
Well, if bringing up new arguments implies that the original arguments are weak or no longer apply, that means that you can dismiss the original arguments – even from a pure logic point of view. So you would have to base the decision only on the merit of the new arguments. On the other hand, if the original arguments are reinforced by the new points brought up, then they provide valid support.
Peter, I don’t think you’ve really met my point on Q5, which refers to the case when there are multiple arguments presented in support of a claim, but they imply different versions of the claim. To add an Australian version, some arguments for a republic implied the proposition
“We should have a republic with an appointed president”
while others implied
“We should have a republic with an elected president”
and it would be perfectly possible to imagine arguments supporting
“We should have a republic with no president”
With these variants, it is possible to produce directly contradictory (and relevant) arguments, both supporting a republic. For example, those supporting an appointed president generally saw a republic as good because it enhanced the power of the (indirectly) elected PM, while those supporting a directly elected president saw it as desirable because it reduced the power of the PM
The use of many (including incompatible) arguments is perhaps related to their persuasiveness to different audiences. As I commented on 17 and 18 June (under the post “Experts and Interests” of 13/6), arguments (perhaps judgements is a better word) which are of less than “beyond reasonable doubt” strength may lack persuasiveness to those who don’t share relevant attitudes and beliefs with the persuader.
An unscrupulous persuader may therefore use many different arguments in an effort to persuade different people. How unscrupulous? Grade 1 unscrupulous would be a persuader who didn’t believe any of them but will gain a personal advantage from successful persuasion. Grade 2 unscrupulous is the same as Grade 1 but where the persuader gains no personal advantage. Grade 3 unscrupulous would be a persuader who believed one argument and also believed that the issue was important enough to use other arguments in which he/she doesn’t believe (perhaps we should have Grades 3A and 3B for those who do and don’t derive personal advantage)…
John,
I may completely have the wrong end of the stick here but doesn’t the UN Charter give you an example of why war may be a special case by establishing a regime that attempts to regulate this type of situation by proscribing the use of force except in “self defence” or by Security Council mandate in order to protect from the “scourge of war”?
Re question 1: according to the Charter, war is special because of the “untold sorrow” it can cause and can only be resorted to in special circumstances and for particular reasons. Moreover, it was drafted with the clear memory of the mutlifarious flimsy reasons to go to war. I’m thinking of WWI and its aftermath here, of which WWII was their final effect.
Re q2: hence the stringent controls in the Charter on the use of force. I suppose loosely analogous to the relatively high burden of proof in criminal cases. In this context, a better analogy is to the general criminal defence of self defence.
Re q3: I’m not sure what you mean by an “LCD solution” but isn’t war a classic case of a non-co-operative solution in game theory?
John —
In my own work (jointly with Katie Atkinson and Trevor Bench-Capon), we have been trying to formalize arguments over proposals for action (in order that machines may engage automatically in such arguments). Our approach represents proposals for action as follows (– I can’t do the symbols here, so will have to put into words):
“Action A is proposed because it will take us from current world-state R to new world-state S, and, in this new world-state S, proposition G is true, and G promotes desired value V (or demotes undesired value W).”
Proposition G may be a compound formula. This formalism separates the objective consequences G (objective at least in principle) of an action, from their subjective valuation, V (or W).
(As an aside, we have catalogued all the ways in which such a proposal for action may be attacked or challenged, and formalized the resulting interaction between a proponent and an opponent, sufficient for machine argument.)
Using this formal representation, your examples (Iraq, Oz Republic) involve different proposals for action, since A, S, G and/or V (or W) are not all the same in every case. What you are asking is how we may compare such different arguments, and can we (and if so, how can we) construct a meta-argument over them all.
Your post and subsequent comments raise some interesting issues, which I will need to think about w.r.t. our formal model.
JQ —
Another potential resource for these issues is the Evidence Project, a large, multi-disciplinary research project led by Phil Dawid, Professor Statistics at University College, London, and funded by the Leverhulme Trust.
However, the project has only been going a short while, and results are still to arrive.
I think that, as far as war is concerned, Gaby has exactly the right end of the stick. Why do we keep forgetting the UN Charter?
I think the right answer is something close to the UN Charter, but
(1) How do we avoid the extreme version of ‘non-intervention in internal affairs’ being pushed by China and Russia right now
(2) What is the underlying principle from which we derive the Charter and which would help us to see what (if any) changes are needed
I think extreme non-intervention in internal affairs combined with extreme ban on external aggressions would be an excellent start.
And, as Gaby said, the underlying principle is right there at the beginning of the Charter – …to save succeeding generations from the scourge of war… The rest is secondary and far less significant.
“Treason doth never prosper: what’s the reason? For if it prosper, none dare call it treason.”
Having donned their rose-coloured glasses, the various interests that coalesced within the Bush clique to manufacture serial pretexts for invasion and occupation of Iraq believed that none would dare call their success mendacity. [These various interests include: neocons and their expansive dreams of Pax Americana; Cheney and crony capitalism; Powell and his tragic inability to escape the role of Step’n’fetchit; Rice and her desperate desire to belong, really belong].
Having patently been raped by reality, the Bush clique, or what remains of it, is now very vulnerable to charges of mendacity.
Historians, of course will be very profitably employed in future in tracing the interactions between self-delusion and ambition as exemplified in this textbook case of maladministration in the rush to war.
However, of more immediate interest, I believe, is discussion tracing the interaction of anguish and panic that will drive the “vietnamization” of post-withdrawal Iraq.