Why Do Farmers Disadopt Successful Innovations? Socio-Ecological Niches and Rice Intensification

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear authors,

I was pleased to see this manuscript, which I have read with interest. I find that your paper makes a useful, original, and relevant contribution to the literature on SRI and on the adoption of new agricultural methods / technologies. It is well written overall and should be received with interest by readers.

I would like to suggest that you consider the following points, which could help you to improve the paper and ensure that it has the best impact.

My main suggestion is that you might make your findings, which are about mechanisms leading to disadoption, more vivid, and potentially generalisable and testable, if you were to organise them into a typology or a set of pathways, and present them as such in summary format, e.g. in a table, a matrix, a flow diagram, or another graphic representation. To make that work, you might consider reframing the presentation of your findings as tests of a set of hypotheses, perhaps based on a theory of change. I believe you are already in a position to frame such a theory, so I don’t think this suggested reframing would require a radical restructuring or re-writing of the paper.

Adding this suggested layer to your analysis might make the paper a little bit longer, and I don’t know how close you are to the journal’s word limit. But I think the work could be done without lengthening the manuscript significantly—a good graphic could convey a lot of information with a few words.

I think it would be open to you reasonably to reject the above suggestion, if—based on your empirical data—you think that it is not possible to develop a typology along the lines I’ve suggested, e.g. because there are too many simultaneous and overlapping causes leading to one general outcome (disadoption), or because each case (i.e. each farmer’s decision-making pathway towards disadoption) is unique/idiosyncratic, so that few general patterns emerge. But it seems clear from your paper that there were some systematic differences between larger/richer and smaller/poorer farmers, and these might be drawn out via a different presentation of your data.

My remaining suggestions are more minor:

Socio-ecological niche is an important concept for the paper. It would be good to provide a concise definition of an SEN and a very brief overview of how the concept has been applied to agricultural technology adoption. You could perhaps expand on the idea of ‘fit’ as discussed by Descheemaeker and others.

I suggest you might flesh out your analytical methodology a little more. For instance, how were the interviews and FGD records analysed? Did you carry out some kind of formal coding? If so, what method or software was used? If you used formal coding, did you adopt a mechanism to check for robustness of the coding (e.g. coding by different researchers)?

I think that a brief ethical statement/commentary would be appropriate, as well as a brief reflection on the social and political economic context for the research, e.g. the fact that you end up with some rather unwelcome findings for the promoters/implementers of the Jai-SRI programme. Did you reflect on potential problems of bias in the farmers’ responses? Did you reflect on whether your report of almost universal disadoption could create difficulties for stakeholders in their relationships with one another, with funders, or policy makers? Did you consider anonymising the description of the project (i.e. not mentioning Jai-SRI by name)?

In lines 199—203, you mention that farmers perceived that they had saved water with SRI, but they were unable to quantify or otherwise substantiate this perception. This led me to wonder whether you ought to make a distinction between (1) the observable and readily measurable experience of pumping less water and (2) saving water (i.e. conserving water resources, e.g water tables), which is less readily observable without specialised equipment and resources. In your interviews and FGDs, was it the case that the farmers were conscious of pumping water less often/for less time, and/or holding less water in the fields, but had no oversight of whether meaningful quantities of water were actually being conserved? My point is that you might elaborate a little to clarify and interpret this finding and what it means.

My remaining comments are minor issues of clarity etc.:

Line 110: ‘principals’ should be ‘principles’

Lines 123—124: ‘…where it could help a … populations’ – singular or plural?

Lines 125—126 ‘Real world applications of agricultural innovations, however, are infinitely more complex’ – More complex than what? Compositionally, the implication isn’t as clear as you assume. Also, is it accurate to say ‘infinitely,’ or is that hyperbole?

Lines 145—146: ‘Of these 100 farmers, we were able to approach 48 of the farmers’ – repetition of ‘of the[se] farmers’ could be eliminated.

Line 147 ‘>2.5 acres,’ apparently this is intended to mean ‘less than 2.5 acres’ (according to row 1 in Table 1), if so, then I believe the chevron symbol should be reversed: ‘<2.5 acres’.

Lines 145, 147, 155—156: The number of interviewees is confusing here. In line 147 you state that you interviewed 43 farmers, but in lines you say that you interviewed 38 male farmers and four female group leaders (=42) and you also carried out a FGD with 5 female farmers (=47 if you’re lumping interviews and FGDs together). Were all of these 47 informants among the 48 farmers you approached (line 145)? If so, why was one farmer apparently not involved in either an interview or a FGD?

Line 379: Is it appropriate to describe SRI as ‘a technique’ (singular)?

I hope these comments are helpful. Good luck with the revisions.

Comments on the Quality of English Language

Overall the paper is very well written and clear. Please refer to my general comments for some minor corrections.

Author Response

Reviewer 1:

• I was pleased to see this manuscript, which I have read with interest. I find that your paper makes a useful, original, and relevant contribution to the literature on SRI and on the adoption of new agricultural methods / technologies. It is well written overall and should be received with interest by readers. I would like to suggest that you consider the following points, which could help you to improve the paper and ensure that it has the best impact.

Kind thanks! The paper has benefited from these suggestions.

• make your findings, which are about mechanisms leading to disadoption, more vivid, and potentially generalisable and testable,
  • organise them into a typology or a set of pathways, and present them as such in summary format, e.g. in a table, a matrix, a flow diagram, or another graphic representation.

Excellent suggestion. We have added a diagram into the conclusion that explicitly links the main argument to four generalizable or testable questions. This is aimed to provide extension agencies with a clearer idea of what social contexts they need to address.

• Provide a concise definition of a SocioEcological Niche and a very brief overview of how the concept has been applied to agricultural technology adoption. You could perhaps expand on the idea of ‘fit’ as discussed by Descheemaeker and others.

We’ve added to this section and provided a specific example of what constitutes a socio-ecological niche in the study area.

• Flesh out your analytical methodology a little more.

We added some minor extra detail to flesh out how the interviews and FGD records were analysed.

• A brief ethical statement/commentary would be appropriate, as well as a brief reflection on the social and political economic context for the research, e.g. the fact that you end up with some rather unwelcome findings for the promoters/implementers of the Jai-SRI programme.
  • Did you reflect on potential problems of bias in the farmers’ responses?
  • Did you reflect on whether your report of almost universal disadoption could create difficulties for stakeholders in their relationships with one another, with funders, or policy makers?
  • Did you consider anonymising the description of the project (i.e. not mentioning Jai-SRI by name)?

We have clarified that the study was conducted almost two years after the project concluded so as not to create difficulties between farmers (who remained entirely anonymous) and project authorities (who were no longer working in the region).

• In lines 199—203, you mention that farmers perceived that they had saved water with SRI, but they were unable to quantify or otherwise substantiate this perception. This led me to wonder whether you ought to make a distinction between (1) the observable and readily measurable experience of pumping less water and (2) saving water (i.e. conserving water resources, e.g water tables), which is less readily observable without specialised equipment and resources. In your interviews and FGDs, was it the case that the farmers were conscious of pumping water less often/for less time, and/or holding less water in the fields, but had no oversight of whether meaningful quantities of water were actually being conserved? My point is that you might elaborate a little to clarify and interpret this finding and what it means.

We have clarified this in the text – thanks!

• Minor comments:
  • Line 110: ‘principals’ should be ‘principles’
  • Lines 123—124: ‘…where it could help a … populations’ – singular or plural?
  • Lines 125—126 ‘Real world applications of agricultural innovations, however, are infinitely more complex’ – More complex than what? Compositionally, the implication isn’t as clear as you assume. Also, is it accurate to say ‘infinitely,’ or is that hyperbole?
  • Lines 145—146: ‘Of these 100 farmers, we were able to approach 48 of the farmers’ – repetition of ‘of the[se] farmers’ could be eliminated.
  • Line 147 ‘>2.5 acres,’ apparently this is intended to mean ‘less than 2.5 acres’ (according to row 1 in Table 1), if so, then I believe the chevron symbol should be reversed: ‘<2.5 acres’.
  • Lines 145, 147, 155—156: The number of interviewees is confusing here. In line 147 you state that you interviewed 43 farmers, but in lines you say that you interviewed 38 male farmers and four female group leaders (=42) and you also carried out a FGD with 5 female farmers (=47 if you’re lumping interviews and FGDs together). Were all of these 47 informants among the 48 farmers you approached (line 145)? If so, why was one farmer apparently not involved in either an interview or a FGD?
  • Line 379: Is it appropriate to describe SRI as ‘a technique’ (singular)?

Thanks for catching these. We have amended all of them as suggested.

 

Reviewer 2 Report

Comments and Suggestions for Authors

This is a novel commentary and contribution on technology adoption broadly speaking. The paper is clearly written in a readable style and clearly makes three main points about SRI disadoption but does lack necessary contextualization with the extensive literature on SRI.

The social-agroecological (or social-environmental framework) is generally reasonable but there it is pretty loosely engaged with here beyond the initial discussion in the introduction.

 

My main critique is that the paper lacks context. In particular the paper could use some bolstering is in the literature review/ discussion. There are numerous papers critical of SRI (and some not necessarily critical) that identify barriers. I don't know the literature well but I am familiar with 2/3 of these and so I imagine each of these three main points come up in the literature (labor, water and knowledge networks). A stronger paper would contextualize its findings in those literature and ideally leverage that previous work to make a broader contribution than here are 3 reasons why people dont adopt SRI.

Some specific comments:
Introduction ln 35-6. What about the possibility that it is purely a behavioral phenomenon- lots of people start things they intend to complete but don't even though they know is good for them (buying a gym membership in January).

Background 76. What categories of smallholders can work effectively with SRI under their diverse social and agroecological conditions across distinct locations-- essentially this is an empirical testable question- and something the CA/SRI/agroecology world has been onto-- some situations where this technology is just going to work better.

 

Sample: It seems possible that the farmers were not exactly selected randomly or appropriately and there may be reasons why the process of including them led to dis adoption rather than the SRI practice shortcoming.

You discuss uneven access to inputs (irrigation for example) but it seems like that was not a benefit to farmers. A broad critique in the CA/CSA/SRI etc literature is that they really only benefit certain farmers and farms. The way you present the paper sort of glosses over those nuances- if everyone disadopts which seems to be the case, that I suppose is less important but it still feels missing here. You dig into the different explanations given for disadopting but with near complete dis adoption it raises the question of whether larger institutional or behavioral factors caused this rather than the specific factors that are presumably unique to agroecological contexts.

Author Response

Reviewer 2

• My main critique is that the paper lacks context. In particular the paper could use some bolstering is in the literature review/ discussion. There are numerous papers critical of SRI (and some not necessarily critical) that identify barriers. I don't know the literature well but I am familiar with 2/3 of these and so I imagine each of these three main points come up in the literature (labor, water and knowledge networks).

A fuller set of literature on SRI has been added, emphasising that many studies seek to control for one or other factor, but few seek to engage a range of social factors as important determinants of adoption/disadoption.

• Introduction ln 35-6. What about the possibility that it is purely a behavioral phenomenon- lots of people start things they intend to complete but don't even though they know is good for them (buying a gym membership in January).

Interesting, but we do not consider this a plausible hypothesis as farmers gave us very clear reasons why they rejected using SRI despite having experienced gains from doing so. The purpose of the paper is precisely to explain why behaviour changed, and - according to farmers - this is connected to the risk profile of using SRI as we explain in the paper.

• Background 76. What categories of smallholders can work effectively with SRI under their diverse social and agroecological conditions across distinct locations-- essentially this is an empirical testable question- and something the CA/SRI/agroecology world has been onto-- some situations where this technology is just going to work better.

It is surprising the sheer volume of publications (cited in the paper) produced by the SRI community that simply focus on agronomic criteria and completely ignore social context. This is why we consider this paper to be important.

• Sample: It seems possible that the farmers were not exactly selected randomly or appropriately and there may be reasons why the process of including them led to dis adoption rather than the SRI practice shortcoming. 

The selection was random, shaped by which farmers were available and willing to talk. There was a fair distribution of farmers that matched prevailing distribution of landholdings so it does seem a suitable selection.

• You discuss uneven access to inputs (irrigation for example) but it seems like that was not a benefit to farmers. A broad critique in the CA/CSA/SRI etc literature is that they really only benefit certain farmers and farms. The way you present the paper sort of glosses over those nuances- if everyone disadopts which seems to be the case, that I suppose is less important but it still feels missing here. You dig into the different explanations given for disadopting but with near complete dis adoption it raises the question of whether larger institutional or behavioral factors caused this rather than the specific factors that are presumably unique to agroecological contexts.

We are not quite clear on the point being made here, but we have clarified the argument in appropriate sections to emphasise that the nuance between farmers is precisely what is important. In this respect, different farmers identified different reasons for disadoption, suggesting that there were multiple drivers of disadoption operating simultaneously.