The Chilcot Inquiry

Speaker A: 00:00:00

he decision to invade Iraq in: 2003

Speaker B: 00:00:52

it was six, seven years really before it got published and there was no set deadline so it really went on and on but I think a part of that was the amount of evidence that was gathered. And so Chilcot looked at all the processes underpinning the decision to go to war and from things like how the intelligence assessments about WMD emerged which is something that was covered in the Butler Review slightly before the Chilcot Report but also other stuff, not just looking at the intelligence analysis but looking at how the government decision processes were working and that there's quite a clear picture that emerges of it being basically a mess that you know the decision to invade sort of happened it sort of washed over people and gradually gradually became the thing we were going to do without anyone really explicitly sitting down and saying well hang on a minute should we should we do this is it a good idea what might happen what are our plans if things start to go wrong and that's where Chilcot is most critical saying that there really wasn't so he said he made statements like the importance of collective ministerial discussion encourages frank and informed debate and challenge you know you need to assess risks weigh options and set an achievable and realistic strategy and he sort of said none of that was really done uh and he said the the lesson really is that all all aspects of any intervention need to be calculated debated and challenged with the utmost rigor now as i i was actually having to be in government i was working in government i was in the ministry of defense not not i have to say part of any of the planning um around uh around the iraq war but uh but i was there and and you know and i've been i've been a civil servant for nearly 15 years and um that is pretty accurate i mean i mean these things it is quite hard to pinpoint okay this is the person that made a decision and here is the evidence on which their decision rests here are the values that they are trying to achieve here are the outcomes they want and now you know here's here's why we believe this particular course of action will achieve those outcomes optimally a decision theorist's approach to how decisions ought to be done would seek to find that somewhere that is not how it and uh i mean i think in the case of the invasion of iraq you know it's debatable whether doing that would have made any difference um but in general um you know government decision making it does not look very much like what you would hope an idealized decision process would look and

Speaker A: 00:03:26

chilcot very much picked up on that okay really good summary um i've already got some questions i want to ask but peter at this stage anything you want to say i think it's it's it's a example

Speaker C: 00:03:38

of something you often see with organizational decision making if you consider the organization as a an entity like a being a conscious being that's making decisions to to fulfill certain outcomes often decisions can't be pinpointed to a particular person making a decision but they are sort of an aggregation of lots and lots of little decisions made by hundreds of people across the organization so down to the level of oh should i include this particular point in my brief should i uh go collect a bit more data about this thing decisions really low level decisions like that have a big outcome later on when it's all this information is aggregated and presents to people who who nominally have the responsibility for deciding whether or not to invade uh these things are clouded things are missed opportunities are lost um and uh it's it's interesting dimension to collective decision making okay so i mean you've kind of preempted the

Speaker A: 00:04:40

the question i wanted to ask but you're kind of getting there um which is is this business on as as usual um and or was there anything particular in this in this set of circumstances the that was that was unusual i guess in this less than optimal um process of decision making or was this just like any other war that goes wrong um any any yeah well let's let's just clarify goes

Speaker B: 00:05:09

wrong because uh you know the it achieved certainly achieved its uh military objectives um in military terms the iraq war was a raging success of course it's the is what happened afterwards it all went wrong it's the political side um where you know that and there really wasn't i mean there were no um you know plans really it was it was kind of made up uh i think people were just hugely optimistic about how stable iraq would be um and and then you know our policy about what to do to stabilize things when they started going wrong was very much made up on the fly and i can i do you know i was slightly party to some of the uh discussions about post-war economic planning um and you know sort of down to things like well which money are they going to use uh are they going to use the old saddam um dinars or do we want to get rid of those and replace them with uh you know the the another currency called the swiss dinar or you know are we going to just have dollars and um you know there there were discussions but no real conclusions and in the end you know it things just happened people brought dollars in and and you know that that that was that was how the decision was made there wasn't there wasn't a sort of weighing up of of uh costs and benefits

Speaker A: 00:06:19

really um so um okay so it was as you say so okay not necessarily going wrong as i said i mean bits went right bits went wrong okay yeah so so my question of of you know was this business as

Speaker B: 00:06:33

normal yeah yeah i'm not special in this well i don't i think yes i mean you see you could say that we were just that what happened in iraq was unlucky now obviously it's very easy with hindsight to say um you know kind of clearly we were clearly unprepared and we should have foreseen um some of that chaos and and it was you know some of the um released jik papers the joint intelligence committee papers uh point to you know all of these tensions and things that saddam uh kept a lid on uh being things that we should worry about um so so you know it might just i think i think it's probably fair to say that this wasn't spectacularly different from from how most decisions are made um but when it comes to these big military decisions there's there's for some reason a much lower standard of evidence um applied generally in government if you look at the decision to uh you know for hs2 so high speed the high speed rail line there is a massive economic assessments there's tons of data gathered um you know there's still debates about whether or not it's correct um that you you know it's years of work go into planning these things and making these cost benefit uh assessments when it comes to going to war typically it's all quite spur of the moment well why is that do you think i that's a very good question and i i think not if i could hypothesize if i could if i could justify that it will be based on the fact that actually when it comes to going to war it's very very hard to forecast anything and if you think about things like how hard it is to forecast the u.s election now we we think the u.s election is going to be fairly important i mean it's going to be quite a different world if trump gets in than if clinton gets in yeah that's peacetime america with tons of data right and we still don't know we still can't really be that sure what's going to happen when you're talking about the aftermath of uh when you invade a country that uncertainty is just you know becomes becomes um uh you know you've got to manage it in a different way and and i think you know most most of the time we don't really know what's going to happen and making it up as you go along which is a classic sort of british approach often does work yeah um in situations where there's very high uncertainty

Speaker A: 00:08:46

but isn't it also that it's if it's a decision to go if it's something like war um it's just a potentially more emotional um looser kind of situation where uh politicians rather than um a bureaucratic uh infrastructure come to the form or they've got they've got a greater role to play uh do you mean don't you think that's what it could be so something like um hs2 is of course i mean the decisions that are made are at a political level but it's it's a it's a different prospect it's a different kind of project than going to war and so i'm not explaining myself

Speaker B: 00:09:29

very clearly no but i don't think i'm yeah you will maybe i mean the thing is that obviously people actually i would guess right my guess is maybe a bit cynical but i think people are probably more emotionally attached to the outcomes of hs2 if they've got a train line that's going to go through their back garden um than anyone was about iraq i think no but the the people making

Speaker A: 00:09:48

the decisions can be more emotion i don't necessarily mean people affected by i mean

Speaker C: 00:09:52

the people i think maybe i think maybe you're right i think emotion does come a big part of it but that doesn't necessarily mean that you can't form a utilitarian argument with emotional and humanitarian objectives if you could if you can evidence that by going to iraq we will um save lots of iraqi lives we'll save lots of european lives because it won't there won't be an opportunity for uh we'll stabilize it and there won't be opportunity for terrorism etc etc then you can form a utilitarian argument but with objectives that are less monetary or less economic so i don't think that's and decision making government is all is sort of has to be but it has to be utilitarian because it's all about trying to uh get the most out of the taxpayers pound to deliver the most possible benefit okay sorry good good decision making doesn't mean

Speaker B: 00:10:46

ignoring values i mean okay look if you're a medieval you're a medieval you know king um and your aim is to expand your empire yeah you can do that in a way that is good decision making or bad decision making good decision making is merely about how you go about achieving your objectives it's not telling you what those objectives should be okay but look now i'm sort of

Speaker A: 00:11:05

talking about it in a different way okay so let's say that um and i want to go on to other stuff with this but let's say that if you can apportion a percentage awaiting to um the power that certain individuals or institutions have within certain decision making processes right i think that something like i think foreign foreign policy foreign policy but actually more used sorry not foreign policy but um the decision to go to war okay military action is often historically um more centered on decisions by smaller and smaller groups of people okay and smaller groups and and the leadership so for example the decisions go to to war in iraq if you could put a percentage on the weight of george bush's um power or what's the word his will to go you're right yeah

Speaker B: 00:12:00

but you're already you're you're abstracting when you say george bush yeah it's implausible to me that george bush uh really really kind of wanted the war in iraq and pushed for it to the to the degree that that sort of simplistic understanding of how government works would suggest i mean it's going to be the case that people in the u.s administration um you know felt that this was a good opportunity to do something that there was sort of more of a mandate in a way you know post 9-911 now the connection between that and iraq is tenuous uh at best and but you know they might have felt well here's an opportunity to get rid of someone who's been a thorn in our side for quite a while um you know i might have side benefits might be might help stabilize the middle east you know but there's all sorts of things that would have would have come together they would have coalesced into a sort of actually this looks like it might it might actually be doable is it a good idea um it was not going to be george bush sitting waking up one morning and saying you know what i'd really like to invade iraq and so you know but ultimately he's the

Speaker A: 00:13:02

person who decides at the end of the day and i appreciate he's got all these sort of uh institutions and processes at his hand to to help him with that decision but ultimately if he didn't want to do it um it it it probably wouldn't have happened yeah well yes if he was ideologically

Speaker B: 00:13:19

opposed to it then then that is uh plausible okay but that's very unusual a very unusual situation i'll tell you just one anecdote from when i was in um mod we were i was taking part in a thing called the the evidence-based decision making initiative and as part of that they tried to analyze uh what kinds of things were sent up to the uh defense council which is the sort of senior decision making body in in defense and when they looked at something you know there's 30 or 40 um papers that asked for a decision they were all they were all to approve something so it was like here's an uh i think with perhaps one or two exceptions where they were actually asked to make a decision

Speaker A: 00:14:01

okay so i take your point because i think the point i was trying to make is the and this is where i want to move it forward which is okay this is what went wrong and it was suboptimal and but looking ahead you know this is how could it be different right and i guess one of the things i'm trying to say it's the human factor i i think um that's what i'm trying to put forward but one of the things you're saying there is okay it might be a human factor but actually you can analyze and and how how information was presented to decision makers and and so actually it is still the the institution or the infrastructure where which is suboptimal okay so um so my question is this how do you stop this happening happening again

Speaker C: 00:14:45

i think there were there was lots of good analysis that went on uh and i think um not not least uh not not least dr david kelly was a big uh it was a big campaigner against um to the point where he sort of actually talked to the press and talked about his misgivings um and uh so there was lots of there was lots of dissent there was lots of people saying well actually this doesn't add up this some of these people had lots of oversight of over lots of different broad sources and say well this this web wmd thing doesn't add up um and this was a big bone of contention when nick and i were both in government that it was it was it was the defense side of things which was sort of arguing against it saying well actually if you consider more things all at once it doesn't seem to make sense but it was the other agencies who are really pushing um the the but the it's the way that this evidence was reported um to decision makers to the 220 blair um and the the the the way bits were cherry picked in order to add emphasis to the argument that's probably the most dangerous thing and that's one of the findings of the report an example is where uh sir david dear dear love presented particular reports that came from a single source about the iraqi capacity for a w launch wmd which later turned out to be an unreliable source okay so again it's fine so how do you stop that happening well i think that we nick and i have talked i've talked at length about about system systemic mechanisms that can be put in place for laying bare all evidence that supports or undermines hypotheses um relying on relying on the computers to do much of the aggregation for you to present you with a much more holistic view of all the data that you've got uh and if it's done if it's done um autumn it done semi-automatically like that it allows you then to do things which you can't currently do uh which such as sensitivity analysis so if you you took that one that one report from this dodgy source you can ask the system how sensitive is it to this the sense the argument to go to war how sensitive is it to this report so if you assume that report is no good then you see that your uh argument to go sort of starts to fall apart so that that that prompts you to go and either collect more data on that or look for other means of supporting the

Speaker B: 00:17:23

argument that's there's that's a i think that well i obviously agree with uh this the vision but even in the short term absent any system like that there are lots of quick fixes that we can put into place and which i've seen actually work i mean i've i've seen um you know when i was in uh i was involved in projects which um did change the prevailing opinion um which i can't really mention but but they they you know there were there were was certain views about for example what would happen in syria that when they're explicitly subject to um to attempting to widen the range of potential scenarios that we were planning on when we were trying to consider a wide range of potential outcomes and potential hypotheses that when people were were sort of made to sit down and and think of options that they hadn't perhaps thought and thought about you know made sure that they're not just fighting the last war um actually the picture did change you know it didn't it didn't it wasn't always a kind of there wasn't always a moment of revelation real life intelligence assessments um where you know the the by sitting down and explicitly laying bare what evidence we had and why we believe the things we believed um you you you can circumvent a lot of the cognitive biases so so you know group think and confirmation bias and so on which which um are inevitable whenever you have a bunch of humans looking at something together um that's a very quick fix you know it's just a matter of saying look what is what are the arguments for and against here how much do we believe them um you know how how important are they to the argument that is not you don't need a high-tech solution you can do that with pen and paper um and it you know it stops it being your view against my view and it starts being a discussion about what the evidence suggests there's no reason you can't do that with a decision to go to war there's no reason at all it's not like somehow that's that's suddenly the principles of good decision making go out the window now there's going to be a lot more uncertainty but that's fine you can factor that in and you can factor that into the kinds of decisions you make so um you know you might if there's if you're very the more certain you are about the outcomes the more you can specialize and optimize your approach so if you're going on holiday to the sahara and you know you could be 99.9 sure that it isn't going to rain you can just choose not to take your umbrella but if you're going somewhere where there's more uncertainty about the weather you're going to want to pack a wider range of tools and and it's no different you know if you've got a high if you have a high uncertainty about something like like what's going to happen to post-invasion iraq um you need a wider range of um political tools at your disposal and i you know chilcot's fairly clear that they didn't have that or at least you know they had to kind of start scrabbling around for them when things started not going according to

Speaker A: 00:20:10

how people wanted them to okay great uh i want to wrap up there is there any sort of anything you want to finally say in one sort of sentence at all peter just to draw the parallel again between

Speaker C: 00:20:24

organizations and individuals or uh or specific examples specific people the uh the the planning bias the the the the the optimism about what would happen post iraq is very similar to the planning bias that individuals or teams suffer from and i i think that much of the problems to do with the way that intelligence was reported to up the chain and it seemed to strengthen arguments rather than counter them um is is a form of confirmation bias but at an organizational level rather than

Speaker B: 00:20:58

an individual level okay good yeah okay nick one final thing yeah just to say that you know uh i suppose not not really to contradict because it chilcot looked not didn't look at outcomes he looked at processes you know actually it's not it's not well things went wrong therefore it's a bad decision things going wrong doesn't mean you made a bad decision uh it might as you said mean that you were unlucky um and if you look at if you look back at the conflicts that the uk has been involved with since the second world war uh i count 32 of which um according to wikipedia at least 19 have been victories now some of those are questionable because they're military victories like iraq afghanistan and libya uh but not necessarily uh you know political or strategic victories um but you know what that means is we roughly we've got a we've got a hit rate of about 50 percent you know 50 percent of uh the conflicts we um get involved with end you know like sierra leone or forklands they they they end uh you know in a sort of you know with us meeting our military objectives i you know what there's a debate about what the right what that figure ought to be you'd like it to be 100 but actually only getting involved with a military intervention when you're 100 sure victory may very well not be uh the the optimal uh figure figure there it could you could well you know it could just be that then you'd be ultra conservative um and you you wouldn't be

Speaker A: 00:22:21

taking gambles that actually were optimal there'd be an opportunity loss um okay good all right thanks guys um well look what i take from this is um according to what you're saying on about 300 or so days of the year i should take my umbrella out with me if i live in england um and be prepared right i think that's that's that's the nugget that i i think i get from this

Speaker C: 00:22:44

so um it depends on how much you care about getting wet i mean i don't mind getting a little bit

Speaker B: 00:22:50

soggy so i never carry an umbrella i i actually have a i've got a waterproof coat i can't stand umbrellas i can't send other people's umbrellas people are very thoughtless when it comes to umbrellas and most people are shorter than me so that's really annoying um but also they're you know you lose a hand uh effectively you've got an umbrella and i i've you know i've raincoats fantastically you wear them and you know the water can't even get in sideways like you can with an umbrella well it just seems like it strictly dominates having an umbrella i don't know why people have umbrellas when they could have waterproofs but i mean so to summarize what

Speaker A: 00:23:25

peter's saying is you've got to be clear in your objectives really what you want yeah um if you don't mind getting soggy that's okay but i would take issue with what you're saying because one of the problems with raincoats unless you have a huge like one of those bizarre australian sort of a flashes coat yeah with people wear those silly hats on and stuff i mean you're going to get wet legs that's my problem you're going to get wet legs with an umbrella no not

Speaker B: 00:23:48

if it's big enough okay anyway no but i think i think the key there is really is uh you know make sure you've considered all the options you know exactly you take you say well look here's my raincoat it's a bit lighter it's easier to wear i get a spare hand but possibly wet legs and then you compare those you know with how much you care about those outcomes yeah it's true

Speaker A: 00:24:07

actually because actually i've got a series of uh quite disposable small umbrellas and what i don't like about them is actually if i'm honest my legs do get wet because the surface area is not quite big enough but i've decided to keep going with that because i don't have a great big long umbrella that i can't fold down because it just gets in the way of things so there you go i'm already sort of doing really i'm using aleph insights um techniques and analysis already right so that's good um okay i'll send you the invoice later thank you okay um thank you very much i'm fraser mcgrew we've been here with nick hair and peter coghill of aleph insights thank you for joining us on the cognitive engineering podcast until next time thank you goodbye