POPREBEL Ethnography Team Call Notes

[amelia]

Team Call Notes 15 Jan 2020

Present: @rebelethno

We mostly discussed @Wojt and @Jan’s analysis framework:

@Wojt

Jan: first draft 11 January 2021
Discussed with Wojt on a few occasions.

1. Culture is a massive (and internally incoherent with occasional contradictions) regulatory system (vide: Russian semiotics, Lotman, Uspenskiy et al). It is thus also an information system, providing answers to four questions. Such answers are often provided in the form of more or less coherent stories/narratives. The questions they answer are:
   1.1. What is it? (meaning)
   1.2. Is it good or bad? (value)
   1.3. How should I feel about it? (emotional cue)
   1.4. What should I do? (script or norm)
2. Types of stories:
   2.1. Everyday stories: close to praxis, common sense, habits (often unreflectively accepted, fragmented, embedded in proverbs, everyday figures of speech, etc.)
   2.2. Myths and ideologies: explicit and elaborate, comprehensive, more coherent and explicitly structured than common sense (that does constitute its own cultural system, as per Clifford Geertz).
3. Ideology: definitions:
   3.1. Basic positions (after Paul Ricoeur):
   3.1.1. Marx: ideology as distortion (for Marx ideological “delusion” is the reason that people do not fight)
   3.1.2. Weber: ideology as legitimation
   3.1.3. Geertz: ideology as integration
   3.2. Ideologies (in political science) are usefully defined as stories (systems of ideas or configurations of political concepts) that: (1) strive to develop explicit, elaborate, comprehensive, and coherent depictions of the world, (2) justify a specific configuration of power, and (3) provide blueprints (scripts, norms) for sustaining or changing the political system and/or changing the world.
   3.3. For example: M. Freeden (tradition of British analytical philosophy, thus focus on language): “ideologies are configurations of political concepts such as liberty, democracy, justice, and nationhood in which particular interpretations of each constituent concept have been selected out of an indeterminate range of meanings they may signify” (1998:749).
4. Ideologies are fully-fledged (“thick”) or partial (“thin”). Thick ideologies provide comprehensive answers to all four questions. Populism is a thin ideology.
5. Typologies of ideologies: a triangle versus a line (liberal conservative, prevalent in the US):
   5.1. Liberalism (dominant value: liberty)
   5.2. Socialism or social-democracy (equality)
   5.3. Conservatism (tradition, community, hierarchy)

So, for coding (my proposals):

1. Values . Let’s not use this term in the following sense: “Peace is a value.” Instead, “peace,” that needs to be first defined (meaning), becomes a value in the following sentence: “Peace is a good thing we need to strive for.” In other words, something becomes a value through the process of (explicit and positive or negative) (e)valuation. In this approach, if someone says: “They have peace over there,” we cannot say/code that for them “peace” is a value. If, however, they say “I want peace” it is a value, however implicitly declared. Why? Because it is hard to imagine that some “wants” something that they consider to be “bad.”
2. Ideology . Let’s code – as much as it is possible – specific ideologies (liberalism, socialism, etc.).
3. Emotions . Let’s use, as much as we can, the typology of emotions proposed by Plutchik (see my note from July 3, 2020), but let’s try to reduce it several simple and basic types. Perhaps 5?
4. Institutions . Let’s follow the definitional tradition that seems to be dominant in today’s social science. Institutions are rules of the game, usually supported by some sanctions. Let me quote myself:

Concrete coding ideas:

1. We may want to distinguish “actors” and “institutions,” simply because some actors are institutionalized, and some are not. For example, “government” (institutionalized) and “demonstrators” or “movement” (not). The point is to catch separately institutionalize and non-institutionalized dimension of politics.
2. Ideally, we would have a coding system in which a post (let’s assume for now that this is a unit of coding; below we introduce the term “thematic unit”) that contains a statement (sentence, several sentences, paragraph) on a specific “object” (say, “hospital”), as well as an evaluation (value) of that object (“bad hospital”) and an emotion (“I am angry because this hospital is so bad,” would be coded three times, with overlapping codes on INSTITUTION (or perhaps ORGANIZATION, given the comment above), VALUE, and EMOTION.
3. We need a decision on how to code emotions. As suggested above, we may want to have a typology of basic emotions, say, 5-10, or simply go for a dichotomy: positive and negative. This should be discussed, and we need a consensus decision soon.
4. The second issue, discussed already in the past months, and which we may want to “close” once and for all, is concerns the difference between “post” and what we will call a “thematic unit. ” Imagine, please the following situation. A post is composed of two paragraphs. In the first paragraph, the author writes about a hospital and expresses negative evaluation as well as anger (emotion). In the second paragraph they write about railroads and they evaluate them positively and express happiness. As a result, our coding of the post as a whole generates:
   4.1. Two objects (coded as “institutions”): “hospital” and “railroad.”
   4.2. Both positive and negative value (evaluation).
   4.3. Two emotions: anger and happiness.
5. As much as we understand, there are two ways of dealing with this undesirable situation:
   5.1. The system knows that codes superimposed on each other need to be treated differently than codes appearing in different parts of the post.
   5.2. The post is broken down into two “thematic units” and this operation two, separate coding clusters that make perfect sense.
6. We may need to go over our coding completed so far and: (1) separate some posts into “thematic units” and (2) (possibly) recode . Given the number of posts already coded, this may be a prohibitively complex and time-consuming exercise. We may want to start, at least, with examining and splitting only those posts which deal with important topics. Below we are beginning to develop a list of such topics. They are chosen due to their centrality in the whole project.
7. Central concepts for which we propose to create “ thematic units :”
   7.1. Actors
   7.1.1. Churches
   7.1.2. Movements
   7.1.3. Governments
   7.1.4. Authorities (local, regional, central)
   7.1.5. The EU
   7.2. Ideology
   7.2.1. Populism
   7.2.2. Conservatism
   7.2.3. Neo-liberalism
   7.2.4. Liberalism
   7.2.5. Feminism
   7.2.6. Anti-communism
   7.2.7. Communism
   7.2.8. Anarchism
   7.2.9. Environemtalism
   7.3. Emotions
   7.3.1. Fear
   7.3.2. Anger
   7.3.3. Distrust
   7.3.4. Unhappiness
   7.3.5. Worry
   The rest is currently being discussed and your input is more than welcome.

Next steps from this meeting:

1. For next meeting, bring a paragraph of text that you want to analyse with the group that is illustrative of any coding decisions that you are struggling with/ that is representative of the kind of analysis you want to do.
2. We’ll decide what categories of codes we want to solidify based on this (e.g. ideology, emotions, institutions, organisations, actions, actors).
3. Keep narrowing down the list of codes we primarily use and merging together codes that have the same or very similar meaning.

Explore abductive reasoning as an analytical frame for the analysis that we do (@Jan can post reading homework).

Discussion

Coding Categories:

Institutions – the set of rules of the game. E.g. healthcare system.

• Things can be more or less institutionalised. Or well-institutionalised (the rules work) vs badly institutionalised (lots of breakdown).
• a social movement may work well but be less institutionalised. A healthcare system can be highly institutionalised but poorly institutionalised (has a lot of rules but it doesn’t work very well).

Organisations – places that have rules, when you walk in you have to follow them (e.g. hospital).

• institutionalised group of actions (social movement theory)

• usually has more resources

• (What is the state?)

• Border

  • not a group of people (so NOT an org), but something highly institutionalised
  • guard, and rules of engagement that if you breach things will go very badly

Actors (human) - e.g. doctors, administrators. What actors make up the organisation?
Actors (non-human) - objects that have meaning, like medicine
Scale - level of affective identifications (do I primarily see myself as a global citizen or as a Polish person, or as someone from the countryside?)

Place/location- as contextual information

–
Goal: break down into a smaller list of codes used more frequently that should be shared. Based upon:

• what came before (first pass coding)
• theoretical interests as narrowed down

Abductive Reasoning

• Peirce
• Read: Abductive Analysis: Theorising Qualitative Research (@alberto)
• Collingwood: the key is the question and how you frame it

Culture both names and evaluates things

• semiotic (meaning)
• axiomatic (value)

–
Discussion about value and how to capture it:

• via SSNA descriptive codes and what they are in a web of associations with, contextual. It’s about the story (e.g. words like fear, uncertainty, dissatisfaction co-occurring with the above categories).
• or impose sentiment analysis layer? hot debate here.

evaluation may be implicit

• key word: “want”

emotional “valence”

• sliding scale colours superimposed upon codes
• to show in forum what is mostly positive and what is mostly negative
• sentiment analysis layer
• how to avoid "anti "+ “not happy” codes

• make a list of emotion codes
• add valence to those codes (add a positive or negative value)
• colour what it touches in the SSNA
  (but lots of disagreement here!)