JOSH: Hello, and welcome to Clearer Thinking with Spencer Greenberg, the podcast about ideas that matter. I'm Josh Castle, the producer of the podcast, and I'm so glad you joined us today. In this episode, Spencer speaks with Robert Chambers about the field of development and Rapid Rural Appraisal.

SPENCER: Robert, welcome.

ROBERT: Thank you.

SPENCER: I know that you've had a long and illustrious career in the field of development. And I want to get a bit with you about some of the insights you've had around how to make that field better. But let's start with a really basic question. So what is the field of development? And could you also tell us a bit about your role in it?

ROBERT: Well, the field of development has evolved. In relation to the way in which the word was used, it was about relationships between donors and recipients, a lot of it associated with aid or with relations between people from the so-called developed world and the so-called developing world. And if you want to know what development is, my reply to that will sound a bit evasive, but it isn't really. Development, for me, is good change. In other words, if something is good, and it involves change, then you can call it development.

SPENCER: Oh, that makes sense. So can you tell us a bit about your role in it? I know you've had a really long history in development.

ROBERT: It began in Kenya, where I was a member of what was then called Her Majesty's Overseas Civil Service, which was the old colonial service. And I was working as a district officer — that's a member of the administration — in rural areas in Kenya. And it was at a time when Kenya was moving towards independence; it was immensely exciting. There was tremendous energy, enthusiasm, and excitement at what was about to happen. That's to say that Kenya was to become independent. And I went on working in Kenya through independence, and working at the Kenya Institute of Administration, training Kenyans to take over the administration. That's how I got into it.

SPENCER: Was development a big thing happening then at the time?

ROBERT: Well, yes. Everything, in a way, was to do with development because we had this concept of developing countries. So everything that happened in those countries, or almost everything, could be labeled as development. It's a very loose general term, but it was used very widely to describe the sector.

SPENCER: Can you tell us what is Participatory Rural Appraisal and Rapid Rural Appraisal, and why did you develop those ideas?

ROBERT: Well, this has been an interest that I've had in methodology, and in how we find out about things, and how we learn about things, which I've had for a long time. And what happened was that many other people got very frustrated with the two main conventional ways. The two methods were social anthropology, which could take a year or more than a year of immersion in communities in the way in which it was practiced. And questionnaire surveys, which had many, many disadvantages in those days, particularly because this was before computers. And so processing and analyzing and using data from questionnaire surveys was very problematic. So I and many others were looking for alternative methods, which we called Rapid Rural Appraisal. The ‘rapid' was the idea of not taking the time for total emotions and social anthropological sort, and not relying on questionnaire surveys, which often produced rubbish or distorted information, and were vulnerable to many disadvantages. So our RA (Research Assistant) was putting together and sharing the various methods that we were using. And when we got together — which we did at the Institute of Development Studies in the University of Sussex where I've been attached for a long time — we hosted two workshops in which we shared what we did, and it was just immensely exciting. When people who are heretics in their own organizations all meet together and find that there are heretics in other organizations which have exactly the same problems, it becomes a bit of an epiphany really, and very exciting, and also very reinforcing one's sense that we were doing something sensible.

SPENCER: Now, I guess people who are heretics find each other on the internet. Back then, it was probably a lot harder to find each other.

ROBERT: Well, it would be different nowadays. But of course, we didn't have the internet then.

SPENCER: So can you tell us what is the Rapid Rural Appraisal method? How does that actually work?

ROBERT: Well, there isn't a method. It's a number of different improvisations, really, which are fitted together. There isn't a manual for it. But sort of issues that all methods that were semi-structured interviews, we found we were all using these interviews, which do not follow a questionnaire, but are more open-ended. So that when new issues arise in the conversation, they can be followed up. And the person who is initiating the interview or the discussion is not tied to having to cover particular boxes.

SPENCER: Can you maybe give an example of what might one of these interviews or semi-structured interviews be investigating? Who might you be talking to, and then how might that conversation go?

ROBERT: Or it could be any one of a large number of people, it could be a local leader, it could be someone working in government, it could be someone working in an NGO, it could be a farmer, it could be anybody. And it was used very widely.

SPENCER: And what are some examples of information you might be trying to learn about from talking to these people?

ROBERT: Well, you might be wanting to learn about the crops that they grow. Or you might be wanting to learn about how they treated illnesses, or what their concepts of illnesses were, very often, but not always, of course. It was to do with finding out people who were local residents, rural people, how they construed the world, and what categories they used. And then learning about those categories. For instance, that there might be a disease, and they might have a word for it. But the word that they had for the disease might encompass something different from any equivalent word that we had. So it was finding out about particularly how local people see things. But also, it was very much finding out with people working in government and people working in NGOs, who were often very good informants, and of course, social anthropologists who do know more than they realize. If you're doing some unstructured interview with a social anthropologist, who's been working in the field, it's quite interesting because they don't necessarily think of the same questions that you think of, and they may be quite surprised. They may reveal information that they hadn't realized was significant.

SPENCER: So you call these semi-structured interviews. So what's the structure to it? Would there be sort of guiding questions, but then you're able to kind of go off script as needed?

ROBERT: People vary in the way in which they do it. At its, um, at its vaguest it's just a euphemism for muddling through. But I wouldn't say that there was any particular structure. Very often, you have a little checklist, at least I do. You have a little checklist, but you don't let the checklist dominate too much.

SPENCER: So what was happening due to the reliance on the other methods? So because people were using social anthropology and questionnaire surveys, what sort of things were going wrong in your view or going off track?

ROBERT: Oh, well, social anthropologists were focusing very much on the issues and the concepts and the mindsets and the categories which they had learnt where they came from, from universities, that they came from (part of their university training). So they came to a rural situation with categories, which they then wanted to put ticks against, instead of coming with a completely open mind to see what they found and what turned out to be significant. They didn't have the flexibility because they would have a Theory of Change, maybe they would have a thesis that they had to write very often for a PhD. And that had to be respectable and intelligible, easily intelligible to the people who are going to interview them for their thesis. So they would tend to be conservative and to stick to known and existing categories rather than new ones, and not wanting to take the risk of new ones even when they weren't standing out. So that was something very much to be got over. And I think on the whole, it was quite successful to liberate social anthropologists from the slavery of the fixed categories and concepts that they had been taught.

SPENCER: It's fascinating how when we go into a situation with categories already existing, that can prevent us from seeing new information, because we keep bucketing things based on our pre-existing assumptions about what we're gonna find.

ROBERT: Yes, that's right. I can give you an example from my own work. I've been extremely fortunate in having funding for field work several times, which was very open ended. And in every case, it's been possible for me to change what I was studying, what I did say I would study. To give you an example, I got a research grant to work in South India and Sri Lanka with colleagues associated with the South Asian Studies Center in Cambridge. And what seemed to be really interesting and important, at a distance, before coming out, was some agricultural research and agricultural extension. When I got there, I discovered that this wasn't people's priority at all. What people, who were small farmers and so on, really cared about and what was really important was water — water for agriculture and water management. And so I was so fortunate, because all I had to do was to write a short note to my supervisor and/or the person directing the project and say, “Look, this is what really matters. No one is studying this. This is the blind spot. And I'm really enthusiastic about getting into it.” And he said, “Okay.” So I was very fortunate to have that flexibility. But biases and blind spots are something that I've considered quite a bit because we need to recognize the biases that we have. We really need to recognize and look for the blind spots, and then do something about them.

SPENCER: I feel that this topic has a lot of generalizability to other things. For example, when scientists are trying to study a phenomenon, they often come in with some idea of what they're looking for, rather than sort of poking at the thing and truly trying to understand it from the ground up. I think a lot of the best scientists do come in with a very open mind, not bringing too many of their own preconceived conceptions. And similarly, when someone's starting a startup (someone's trying to build a new product or service), I think when they go in with a really open minded curiosity, interviewing users of their product, or showing their product to people and getting their reaction, they can learn a ton of things that they didn't realize about how people perceive their product, or whether or not it's solving people's needs. Whereas if they go in thinking, “Oh, yeah, of course, my products are useful.” They often are blinded to the fact that it's not really serving their users. So I just think that this concept goes way beyond development.

ROBERT: Yes. The contrast between really being top down and bottom up to put it very crudely, that if you're top down, you have your concepts. And then you try and fit the reality that you encounter into those pigeon holes. But if you're bottom up, you're actually eliciting, you're finding out about the reality. You're open minded. And what I think everybody, in whatever way of life they're involved in, can realize and experience that it's just bloody good fun to be exploratory. I think the word “explore” is an important one. And the concept of being an explorer — I don't know if you or any of anyone who may be hearing this has ever done that exercise of self-discovery, which is, you draw a circle, and then on the outside of the circle, you draw other circles, and you ask the question, “Who am I?” And you fill in the outside circle until you've completed it. You know, “Who am I? I'm a father, a social scientist,” like that. And then in the middle, you say, “Okay, well, then, who am I really?” I mean, what would I like to define who I am and what I do? And for me, the word explorer came up.

SPENCER: It seems like the mindset of curiosity is also essential to this, right? You're paying close attention mindfully to what's actually going on. And you have this curious mindset, trying to notice what the person is saying, what surprises you and so on, is that right?

ROBERT: Yes, absolutely. And I mean, this goes for field work as well. Noticing things...it is quite amazing how often one doesn't notice things which are significant. Sherlock Holmes hadn't had this. In one of his sayings to Dr. Watson, he said, “You see, but you do not observe.” And ‘observe' was to really mean, think about it and take notice of it. Otherwise, you can just see things. I had this with soil and water conservation works that farmers in South India and in Sri Lanka that they themselves construct. And it was absolutely astonishing to me. Professionals didn't notice what they were doing. They were doing a lot of soil and water conservation works themselves. And you could walk straight past it. I did walk straight past it, to start with. And then I rediscovered it when I went on an early morning jog. And then I realized that there was this whole area, which was indigenous solid water conservation to be opened up and explored. I mean, not explored from the farmer's point of view, but from our point of view.

SPENCER: It seems this has to do with the way the mind just ruthlessly pares down information. If we're walking down the street, 99.99% of stuff that's there, we're not even consciously aware of, right? It's just kind of automatically ignored. Whereas we pay attention to some small fragment of what's there. And by being mindful, you can increase the amount that you're noticing what's going on. I found this very interesting to try to do during conversations where I try to become super aware of the conversation itself. And I noticed that more and more information starts coming to me during the conversations as I kind of take on that sort of more mindful lens.

ROBERT: Yes, I think that's right. In a conversation, you can notice something that another person has said and pick up on that, and explore it, and probe it. And if you get it right, then it can open up a whole field of discussion and discovery.

SPENCER: Right. Even noticing that someone was hesitant about answering or that their tone of voice changed, the subtle cues that there's something else going on here that's not just literally in the words being said, it's so easy to miss, but might be crucial to understanding someone's perspective on something.

ROBERT: Yes, indeed.

[promo]

SPENCER: So going back to the two previous approaches that were used in development, the other one was questionnaire surveys. Could you tell us a bit more about what was going wrong by people relying on questionnaires surveys? What sort of bad outcomes was that producing?

ROBERT: Well, in those days, we didn't have computer processing. And what was happening was that it was taking about eight months to process the questionnaires to add them all up. There was even a time when they were added up by punching cards, which were then in a box and you put a needle through and then lifted it up and saw which cards it brought up. And they wouldn't be the cards in that particular category. That was being done. I did that. And we were just getting over that. But the processing of the data usually took about eight months or so. So it was a very, very slow process compared with what we have nowadays. The other thing that was really very seriously wrong with questionnaire surveys — but there are two things — One was that they had preset categories. They were set at the beginning. And then you had the questionnaire, and then you implemented the questionnaire. And so it was a fixed structure of knowledge and ideas that was being reinforced by the questionnaire. And very often, the questionnaire was set up a long way from where it was going to be used. And even without visiting it, I once took part to my shame in designing a questionnaire for South India and Sri Lanka, sitting in a room in Cambridge in the UK—shameful. [laughs] And I'm appalled to confess that even now. The other thing that was wrong was inaccuracy, that very often the information was distorted by the generality of the relationship between the person asking the questions and the person responding. And there could be distortions in this which were to do with power relations, or with opportunism, or with other motivations. But these could occur across many of the questionnaire, interviews situations. So they could come up with something which correlated but which was misleading. This was very little recognized.

SPENCER: Can you give an example where the power relationships of the dynamic of interviewer and interviewee led to kind of bias in the data?

ROBERT: Well, a person who's answering the questions will be thinking, “What's in this for me?” and may well be quite cynical because of past experiences or just from common sense about the situation. And maybe, for instance, (I mean, in a very crude way), if you ask people, “What do you need?” They'll say money very often. It's that sort of level sometimes. The respondent is trying to figure out what the situation is and what the advantage that can be for her or for him and may massage their answers to fit those expectations, and to try to draw them out.

SPENCER: So is the idea that the person being interviewed, when asked “What do you need?” They might think, “Well, the thing this person is most likely to provide me is money. So I'm gonna save money because that would be nice. And maybe I'll actually get it if I say any money.”

ROBERT: Yes.

SPENCER: So I designed a lot of questionnaires. And one thing I found really useful is, in early draft of the questionnaire, when we're asking people with multiple-choice questions, to then ask the participants to explain their answer afterwards. And this has just been incredibly valuable, where we're often surprised by the explanations people give for why they answered a certain way. Sometimes we find that they're not interpreting the original question the way we thought they would. Sometimes there's just more nuance to their answer than we expected, or the interpretation of their responses is not what we thought. And this really led to us, I think, getting much better at designing surveys, where we can sort of understand both the quantitative answer (what percent answer this way), but also why do people say that. And so I kind of think that quantitative surveys are useful for measurement. But qualitative surveys are really useful for understanding the why behind the measurement or just understanding what you've really measured. And if you just do one or the other, if you just do qualitative, then you may lack measurement. But if you just do quantitative, you may not understand what you've measured. So that's kind of how I come to think about it. I'm curious if you ever reaction to that,

ROBERT: I find the distinction between quantitative and qualitative a bit distorting. This is because it tends to overlook a middle ground. The middle ground, we sometimes call ‘participatory numbers' or ‘participatory statistics', is having facilitating processes which generate numbers. And this can be done in many different ways. But some of the best ones which were associated with PRA (Participatory Rural Appraisal) are physical (that's to say people), play with, let's say, beans, and they can show relative proportions by different piles, and generate numbers. Because it's participatory, it might be regarded as a qualitative way. Maybe this is a quibble about terminology, and unimportant, but I do think it's really important to recognize and to exploit the potential of physical, visible things to generate numbers, or to demonstrate realities in other ways.

SPENCER: Can you give an example of a question that might be asked and using that kind of method?

ROBERT: Well, you could ask people the proportion of different foods which they eat. So they would have all the food in the day in a small pile of beans or something, and then divide it up. And one of the advantages of this is that people argue with one another, because they could see what's being said. This is underestimated. The importance of being able to see something which has been said, and which is not disappearing. Words disappear — maddeningly, as we all know — whereas if you're representing something physically, it doesn't. It stays there.

SPENCER: So it seems like it reduces the kind of working memory requirement because you can kind of look at what you've done, and you can make adjustments. But I imagine, it also might be really useful for people that, let's say, have never had any training in math.

ROBERT: Yes, I suppose. But usually people have a much greater capacity than the investigators recognize, at least in development studies.

SPENCER: Is that a common theme you've seen...sort of underestimating the people being interviewed?

ROBERT: Oh, yes, underestimating what they're capable of doing. We have this phrase — well, it comes from participatory mapping, which maybe we'll discuss presently, but this phrase, “handing over the stick of what you can do, if you imagine we're on a dirt surface.” We want to ask someone about their community, and if they can make a map of their community. And if they say, “We don't know how to make a map. Come on, you do it.” That has happened in the past; it probably wouldn't happen so much now. And so you started, you say, “Well, that's the school's over there. Is that okay? Well, let me just put in a square for the school on the ground. And the road comes like this, doesn't it?” And then they say “Yes”. And then there's a point at which you hand over the stick. And that's what you're working towards, so that they take the stick, and they take over the process and create a map. And people's capacity to do this is far greater than has been generally recognized. Although it's much more recognized now than it was 10 or 20 years ago.

SPENCER: So can you talk about that technique for a moment? I like that idea. I don't know whether you mean it literally or metaphorically. But it seems like something about kind of getting the drawing started, helps the interview subjects understand what your question really means or where you're coming from, and then they can get engaged with it?

ROBERT: I think it's almost always a really good idea to have the idea that the concept that you're working towards, that you're going to hand over the stick, that you're going to start a process of thinking and analysis and of sharing ideas, which they will then themselves facilitate and run, not just you.

SPENCER: Got it. So you mentioned Participatory Rural Appraisal, which is different from what we mentioned earlier, Rapid Rural Appraisal. Could you talk a little bit more about what is Participatory Rural Appraisal?

ROBERT: This is a source of Rapid Appraisal, in which people themselves (rural people) analyze their situation. And it's facilitated in a sense that an outsider may convene the meeting, an outsider may start the conversation going with the objective of handing over. So it's rural, because many of these initiatives were rural. But of course, it can be open as well, and is open as well. And it's participatory in the sense that people take it over and do it themselves. And so the outsider's role as a facilitator is to get all this started and happening. Convening is one part of it. Starting a conversation is another. And perhaps, nudging it in a certain direction, if that's undesired. And then sitting back and shutting up. The problem that we have so often is that we talk about, “I'm going to go and talk to a farmer.” How often have you said that? I don't know how often I've said it. [laughs] But it's a bit shameful because talking to is very different from listening to. But you don't say “I'm gonna go listen to a farmer.” You can do it and I had heard of it, but it's not very usual. We need to train much more in listening and less in talking. And to value listening more, and to value talking less. I was in an immersion (that's saying we're in a community), a short immersion with the person from the German Aid agency who was in charge of recruitment. And we went through this experience of not talking to people, but of just living with them, and maybe asking some questions. And when we had finished all this, she said, “You know, when I'm recruiting people to work in my organization, I have always gotten people who can talk well and judge them on that. And now, I'm going to switch that. And I'm going to try and look for people who are good at listening.”

SPENCER: This reminds me of when I tried to help friends who are dealing with difficult situations. I've come to think over time that actually most of what I should be doing is just asking them questions to help them reflect on their own challenges. And then if I give any kind of advice at all, which often they don't want. But if they do want advice, that has to be given with the knowledge that they know a thousand times more about their situation than I do. And so, I might be putting out an idea, but really, they have to evaluate that idea with respect to all of the information they have about the situation that I'm ignorant of.

ROBERT: That's right. If you're asked questions, and you're an outsider, it's usually possible to turn them around into questions to be explored by the participants. And learning to do that can be, perhaps, quite important.

SPENCER: So another concept you mentioned briefly is Participatory Mapping. Could you tell us what that is?

ROBERT: Yes. That is asking people if they can make a map on the ground. There are methods for doing this. And it all started in South India. And it was a very exciting process. It may have been used elsewhere in the world, but this was seminal. We had done special colored powders called rangoli powder. And these were used together with making marks on the ground, using stones to represent, say, a school or temple or mosque. And people draw on the ground. It can be quite big and be 10, 20, 30 feet across. They draw on the ground a map of their community. And you learn so much from this. And it's possible also to develop it so that every household is shown, and the number of people in the household and so on. So that spot is Participatory Mapping. It can also be done on paper. But the advantage of the ground is that you can alter things and adjust things and correct things. Whereas on paper, you have to run out or cross out, which is all rather messy as you approximate towards an agreed representation. And I think this is an element that's really important in all these processes. And that is the idea that they're evolutionary, that emergence is to be expected and valued, that things come out that you didn't know to ask. And that's part of the excitement. The more participatory you are as a facilitator, the more things will come out, the easier it will be for people to take over the stick.

SPENCER: I imagine that in addition to noticing things that you didn't expect, there's also an element of building trust by showing curiosity and genuine interest and asking questions and not talking too much. Do you think that that builds a sense of trust with the participants?

ROBERT: Yes, I do. Not talking too much, noticing things, asking questions about things that you see that in rural areas can be a very, very significant source of understanding (and in urban as well). Trust is this curiosity that you mentioned: having a curiosity, asking questions, and showing an interest and then to lay just what comes out.

SPENCER: Did you experience distrust from your participants sometimes, or confusion about why you're asking them so many private questions?

ROBERT: No, not much. It depends on who you are, who people think you are, what people think you're doing. If you start any inquiry by explaining all that, and why you're doing it, and what you're going to get out of it or hope to get out of it, so that all the cards are on the table, it helps a great deal. Otherwise, people (and I've had this) become suspicious. “He's doing it because he's doing a PhD. He's not really interested. He's just doing it for his PhD.” I do see some people hearing this would have had that experience themselves.

SPENCER: How does it work to map out the values or needs or priorities of people in these rural areas?

ROBERT: Well, it's not difficult. You can ask people to list the various priorities that they have. And then ask them to sort pieces of paper — maybe you have a piece of paper for each priority, and then you ask them to sort them. And then there's a process called matrix scoring, which is very powerful. And I don't understand why scientists don't use it more. Matrix scoring is quite simple. You have items which are going to be compared, or processes which are going to be compared, or some things which are going to be compared. And you put them across the top. And then down the side, you have indicators of different sorts, so a list of indicators. And that gives you a matrix. You can draw the matrix quite easily. And then you score in each one. And it's great fun. You can do it with beans, or discs or anything. (I happen to like beans.) But if you do that, you won't have it all the same, unless you're wanting to score something quite different. And then there are different ways to do it. (I won't go into the different ways of doing it.) You can invent your own way of doing it. You end up with statistics.

SPENCER: So just to clarify, what sort of indicators might you have? Could you maybe give an example of how you do this?

ROBERT: Supposing you were doing types of food. So you would have across the top: oranges, lemons, bananas, parsnips, potatoes, and so on, and so on (whichever ones you wanted to compare). You'd have that across the top, or it could be different sorts of cooked food. And then down the side, you would have — I have people done this, I've stacked them on my wall, and so on. — taste, cost (those are two important ones), accessibility, trouble in cooking or preparing. And you can make a list like that. And if you make a list like that and start scoring, you'll very often find that there are other categories which emerge in your mind, so that you end up with a longer list of indicators than you expected.

SPENCER: My understanding is that men, women and children may map out and value things differently. In your experience, could you talk a bit about that?

ROBERT: Oh, yes. Well, there are very significant gender differences. One of the advantages of these methods is that if you convene women, all on their own, or children all on their own, or men all on their own, and you ask them, for instance, to do some matrix scoring of — we had the example of food — if we had cooked food, what they liked most and what they liked least and so on. Then you'll find that they're different. And you will find, for instance, perhaps that women who prepare the food have got quite different criteria that men don't think of to do with: how much time it takes to collect fuel it uses, and so on. So the realities and the priorities are different. And for children, things emerge, which otherwise would not have been recognized, for instance, dogs with children, dogs often come up as being a major problem for children, which adults don't recognize.

SPENCER: And I know you've also looked at how these sort of value matrices might differ between Westerners and people in these so-called developing countries. Do you want to comment on that?

ROBERT: I have had the unfortunate experience of in a country in the South, people thinking that I'm sort of acting down to them in a rather patronizing way and asking them to do these physical things, showing scores. That's difficult to deal with. It's not true, but it is true that these letters are not used nearly as much as they might be. And particularly not in universities, which is a very sad tragedy.

[promo]

SPENCER: So you've had a long career in development, and I'm wondering what sort of changes have you witnessed in the field? And do you view those changes mainly as positive? Or do you actually see things that have gotten worse?

ROBERT: Oh, what a big question. I think something which has got better is to do with relationships. When I started in development, it was very much an outsider dominated set of processes that we were talking about. That has changed absolutely radically. And we're now talking about things which are much, much more under the control of people living in developing countries. Their investigations, their analysis, their conclusions, their actions. I think another thing that has changed a lot has been recognizing power as being a factor. We didn't use the word ‘power'. I remember, we were being cautious about using the word ‘power'. When we had people from DFID, from the aid agency, when we had them coming, because they might think that we were getting at them, that was misguided on my part. And ‘power' is now very much on the table. But the other is the role of an investigator, that's now much more the role of a facilitator who is encouraging people to do their own research, their own finding out, and come to their own conclusions.

SPENCER: These seem like positive developments. Are there other developments that you think have been negative, or things have actually gotten worse?

ROBERT: Hmm.. Well, nothing springs to mind. Mmmm.. Yes, I can think of one thing. And that is that some of the earlier innovations, which were really powerful, particularly Participatory Mapping, and recognizing that people have got spatial insights, and that these are significant. I think that that went through its heyday, and then has rather lapsed. I do regret that. And I think also, participation, in general, is not used nearly as much, as would be optimal. And this goes for much more than development. It goes for scientific research. It goes across the board, in terms of who learns from who, who has the analysis. And we had these traces whose knowledge cards, whose ideas, whose concepts, whose indicators, whose language, and all of these are significant. Take the example of language, for instance. Language can be a medium for being very dominating and putting down other people by using words which they're unlikely to know, or unlikely to use with facility. And that is something an exercise of power in the relationship, which I would hope is more recognized than it was before. But it can apply. I can apply in hard sciences. I can be applied to pretty much anywhere. And this, I think, is the importance of recognizing biases and deliberately offsetting them. The biases — I've particularly gone to town where there have been the biases of Rural Appraisal — of going out from an urban center and then stopping and walking around and finding out about things, there are very strong biases built into those processes. And I could talk a considerable length about this, because it occurs so often, and it's so universal. I think anybody at all in any science or any discipline, but particularly in social sciences, needs to be aware of how their behavior has got biases built into it. And so blind spots. And one of the things we're not as good at as we should be, and this applies very much to the sciences in general, is recognizing what we don't know. And recognizing the subjects that we don't study, perhaps because they're difficult to measure, or perhaps because they're unfashionable, because they're difficult to get funding for. And we need to recognize the blind spots. One, for instance, that I mentioned, because I've been involved in a little bit myself, has been the relationship between undernutrition and stunting, and frequently transmitted infections. Amazingly understudied. And their impacts are amazingly understudied, so that many people think that stunted children are like that because of nutrition. It came out in a very recent broadcast, that people talk about the puzzle of why is it that even when nutrition is good, people are still short? Well, the answer is that they have a condition very often called environmental enteropathy, which is having bad bacteria in their stomach, which then gets into their bloodstream, which then has to be fought with continuous use of energy by antibodies. It's a blind spot, and it remains a blind spot. When we look back a hundred years, people hearing this may have had this experience themselves that they go into a farmhouse, which was built 200 years ago, and it's difficult to stand up in the kitchen, or it was difficult. — I've had that experience, and I'm only just over six foot — not being able to stand up in the kitchen. And that's because people were shorter, and they were shorter, not so much, in my view (and I can substantiate this, to some extent) not because of a bad diet. You can change the diet and people will still be short, but because of infections from the feces. And this has to do with lack of sanitation. So when sanitation improves, people are taller. It's as simple as that–sanitation and hygiene.

SPENCER: That's so interesting, because yeah, I had heard it was due to nutrition. So it's fascinating that that might actually be a misconception largely.

ROBERT: Yeah. Well, look, if you're a nutritionist, nutrition is your thing. Often, defecation and fecal transmission of infections are not your thing. It's not a category. It's not an explanatory factor in your mindset. Actually, in my view, it is tragic in this particular case that this has gone unrecognized for so long. And it still is.

SPENCER: The last topic I want to touch on quickly before we wrap up is the idea of start, stumble, and self correct. Could you tell us what that is?

ROBERT: Oh, start, stumble, and self correct. Share is the fourth one. This was something I just concocted. This summarizes a way of going about participation or development, that you start, you launch into something. You don't sit and look at your navel for a long time and ponder. You get down to it. And you learn by doing and by making mistakes. I think learning by making mistakes and rejoicing when there are mistakes to learn from are very important. So that's Start. And then Stumble. Not everything works. You start something. You try it. Not everything works. Self-correct means you, yourself, change direction or correct whatever you're doing. And then Share the outcome with other people. And this is sort of a short mantra. So, one dimension of going about development.

SPENCER: I know that you came up with it in a development context, but it seems to apply to so many things in life where almost anything difficult and bold that we want to do, often it's better to jump into it to notice when you made mistakes and learn from them, self-correct and then share what you learn with others.

ROBERT: Absolutely. I mean, there are things which delay our learning. One of them is funding, spending a lot of time trying to get funding for something instead of just digging in and learning on the run. But you can't do that if you haven't got money. Funders, in my view, need to be much more exploratory in their funding, so that people can do that. They can start, they can self correct, and they can give feedback quickly. And this goes particularly before any large project of any large investment project in a pilot before the main project. So it becomes part, it feeds into a design process. But it is this idea of learning continuously, and not being too preset in any groups. And that seems to me to be important. And also, it can lead to a lot of fun.

SPENCER: Robert, thanks so much for coming on today. It's a really fun conversation.

ROBERT: Well, thank you. I've enjoyed it.

[outro]

JOSH: A listener asks, “Has your mind ever been changed by an interaction on Twitter?”

SPENCER: So my mind is often changed by Twitter, but through a kind of unusual mechanism, which is that I run a lot of Twitter polls and I make large predictions of what I think the percentage breakdown will be in the polls just as I post them. And this gives me an ability to see if I actually predicted correctly. And I'm often surprised by the results. Now that's partly because I'm trying to choose polls that will surprise me because I'm trying to use polls where I don't think I know the answer. And the reason I do this is because I'm trying to improve my predictive ability of predicting what people believe, so that I can actually predict without running a poll, “Oh, people will probably say this.” Of course, there's a demographic question. I do have a decent understanding of the demographics of my Twitter followers. So I try to adjust for that in my mind. But I'll just give you a few examples of Twitter polls that I was quite surprised about. So the first is the following question: Compared to your close friends, do you think the problems/challenges you're currently facing in life are bigger that is more severe, the same size, or smaller than theirs on average? To take a moment to think about what you think people will say to that, that they're either that their problems are bigger than average, the same size, or smaller than the people that they know. So I was surprised to find that 47% of people said that their problems are smaller than average. And that was by far the most common answer. And the other two answers got about 26%. So that quite surprised me. Another Twitter poll I was pretty surprised about was when I asked people, “How is your happiness now compared to what it was five years ago?” And they got to choose either “I'm happier now”, “about the same” or “I'm less happy now”. Take a moment to think about what you think the answer will be. So I guess the most common answer would be about the same. That made intuitive sense to me, but actually, almost 60% said they're happier now, which really quite baffles me, with about 20% saying the other two answers. So I'm a little confused about that. Why is it 60% of people say they're happier now than they were five years ago? I don't get it. It's very interesting. The third one that I'll point to, I asked, “Are the institutions of power in the US reliable/trustworthy/competent 20 years ago? And are they today?” It was sort of a two by two grid of answer. They could say, “then no”, “then yes”, “now no”, “now yes” between “Were these institutions of power in the US reliable 20 years ago, and are they today?” And to my surprise, two thirds of people said, “no, no”. In other words, they don't believe that they were reliable 20 years ago, and they don't believe they're reliable now. And the reason this surprised me, is that I had this perception that a lot of people thought that institutions decayed over the last 20 years, that they used to think they were more reliable, and that they decayed a lot in reliability. But only about 17% of people said that. So that surprised me a bit. I thought more people would have said that.