Literature Research and Analysis

History of Social Security in Switzerland

A guided exercise for systematic literature research and problem-oriented literature analysis with critically controlled use of generative AI.

Author

Affiliation

Moritz Mähr

University of Bern

Overview and Didactic Goal

This exercise introduces systematic literature research and problem-oriented literature analysis – with the goal of being able to reconstruct the state of research on a historical topic. A focus lies on the critically controlled use of generative AI as a heuristic tool for expanding search spaces and structuring texts.

Case Study: History of Social Security in Switzerland

The exercise follows a circular search and analysis process: the starting point is a broad topic area (social security), from which a clear problem focus, a refined research question, and a reproducible corpus of relevant secondary literature are successively developed. Manual research and analysis phases alternate with AI-assisted explorations, which are each critically validated.

Prerequisites

• Basic understanding of historical research methods
• Basic knowledge of working with generative AI (especially prompting)

NotePrompt Engineering

If you are not yet familiar with prompting, we recommend completing the Prompt Engineering exercise first.

NoteLLM

You can complete this exercise with LLMs from different providers. For this exercise, it is helpful if the LLM has internet access and allows file uploads.

Learning Objectives

Upon completion of this exercise, you will be able to:

• justify the function of literature research and analysis in historical research,
• translate a topic on Swiss social security into a researchable search strategy,
• conduct and document systematic (catalogs/databases) and unsystematic (snowballing) research,
• use AI as a heuristic search aid (keyword expansion, query design, screening),
• systematically assess relevance, argumentation, theory, and methods of secondary literature,
• identify central lines of debate and research gaps and derive a refined research question from them,
• implement literature management in a reproducible and citation-secure manner (Zotero/BibTeX, search protocol, decision log).

What is Literature Research and Literature Analysis?

Literature research is the systematic identification of relevant secondary literature (and relevant tertiary resources) to map the state of research.

Literature analysis is the problem-oriented, critical reading of this literature to reconstruct arguments, theoretical references, methods, evidence bases, and controversies (rather than extracting “facts”).

Case Study: History of Social Security in Switzerland (welfare state, AHV/IV/UV/KV/ALV, federalism, associations, gender order, transnational transfers).

NoteNote on Online Research

Despite advancing digitization, historical research cannot be conducted entirely online.(Milligan 2019) In particular, out-of-print secondary literature and sources not yet catalogued often require physical research in libraries and archives. Plan for access (library, interlibrary loan, archive) and note that platforms/catalogs structure visibility (cataloging logic) and do not guarantee completeness.(Putnam 2016)

Structure of the Exercise

Each step contains:

• an objective,
• concrete tasks (incl. optional AI prompts),
• a work and reflection assignment.

1. Problem Formulation and Search Horizon
2. Orientation Knowledge via Tertiary Resources (Debate and Conceptual Framework)
3. Systematic Search I: Catalogs (Coverage and Retrieval Logic)
4. Systematic Search II: Journals/Media and Special Paths (Context, Reviews, Concepts)
5. Snowballing: Backward/Forward Citation (Control Canons and Blind Spots)
6. Screening and Quality Assessment: Relevance, Argument, Method, Evidence
7. Historiographic Debate Mapping and State of Research Text (Condensed, Evidence-Sensitive)
8. Research Gap and Delimitation: From State of Research to Research Question
9. Reproducibility and Documentation

1. Problem Formulation and Search Horizon

Goal

From the broad topic area (social security) to a clear problem focus and an initial search hypothesis.

Task (Without AI)

Formulate in 5–8 sentences:

• Investigation period (e.g., 1880–1950 / 1938–1948 / 1970–2005)
• Subarea (e.g., old-age provision, health insurance, disability, unemployment)
• Presumed conflict or mechanism logic (e.g., federalism/referendum, associations, expertise, gender, transfer)
• Expected literature spaces (history / political science / legal history / social policy)

Task (AI Exploration)

I am working on the topic "History of Social Security in Switzerland."
Suggest 6 problem-oriented, historically workable focus variants (each with: time period, actor/arena focus, mechanism).
For each variant, provide 8 search terms (DE/FR/EN; historically plausible terms) and 3 expected literature types (monograph, article, anthology chapter).
No invented titles.

Critical Check

Mark in the response:

• anachronistic categories (mark as retrospective search anchors),
• implicit theoretical assumptions (e.g., “welfare regime”),
• missing perspectives (e.g., gender, cantonal level, ILO/transfer).

2. Orientation Knowledge via Tertiary Resources (Debate and Conceptual Framework)

Goal

A robust contextual framework (chronology, concepts, institutions) as a basis for targeted literature research.

Core Resources

• Platform:
  • History of Social Security (Switzerland) — Home
  • History of Social Security — Synthesis
  • History of Social Security — Bibliography
• HLS Entry Points:
  • Historical Dictionary of Switzerland: Social Insurance
  • Historical Dictionary of Switzerland: Social Policy
  • Historical Dictionary of Switzerland: Old Age and Survivors’ Insurance (AHV)
  • Historical Dictionary of Switzerland: Social Security

Task (Without AI)

Read (skimming):

• one synthesis stage on the platform,
• one HLS article (or several briefly).

Extract:

• 8–12 key terms (incl. historical terminology),
• 6 “fixed points” (events/laws/institutions),
• 5 “debate markers” (e.g., federalism as veto point, state/private relationship, gender, transfer, expert knowledge).

Task (AI Exploration)

Based on the following terms/fixed points: [insert]
1) Build a debate map (4–6 debate axes) on the history of social security in Switzerland.
2) Assign 5–8 typical author/work types to each axis that one would typically expect (without inventing specific titles).
3) Derive 10 search strings for catalogs (Boolean, phrases).

Critical Check

Validate “debate axes” with at least 2 independent entry sources (HLS/platform/anthology TOCs).

3. Systematic Search I: Catalogs (Coverage and Retrieval Logic)

Goal

Reproducible, documented search in major catalogs; building a starting corpus.

Primary Search Location

• swisscovery (SLSP)

Task (Without AI)

Conduct 6 search runs in swisscovery:

• 2 each with broad terms,
• 2 each with precise phrases/filters,
• 2 each with multilingualism (DE/FR/EN) and name variants.

Document per run:

• Search query (exact), filters, number of hits, export date,
• 3 selection criteria (relevance, time period, publication type),
• 3 exclusion criteria.

Task (AI Exploration)

Conduct the search runs heuristically using AI. Provide sufficient context.

You are a historical research assistant with knowledge of Swiss social, economic, and political history.

Context:
- Research topic: History of Social Security in Switzerland
- Time period: [e.g., 1880–1950]
- Goal: Building an initial, broad starting corpus of secondary literature

Task: Search for relevant secondary literature.

Conditions:
- No invention of specific titles or authors.
- No pretended catalog knowledge.
- Mark uncertainties explicitly.

Critical Check

• No title from AI suggestions is adopted without being found in the catalog.
• For each adopted title: note the “adoption rationale” in 1 sentence.

4. Systematic Search II: Journals/Media and Special Paths (Context, Reviews, Concepts)

Goal

Make debates and state of research visible through journal literature, reviews, periodicals.

Search Locations (Examples)

• Journal platforms: E-Periodica — Swiss Journals Online
• Newspapers (for discourse and reception paths): e-newspaperarchives.ch — Digital Newspaper Archives

Task (Without AI)

Select 2 journal titles or periodical corpora and search for:

• “social insurance,” “AHV,” “disability insurance,” “health insurance,”
• plus your focus term.

Extract:

• 5 relevant articles or reviews,
• recurring terms/frames (e.g., “burdens,” “solidarity,” “abuse,” “three pillars”).

Task (AI Exploration)

From these 5 abstracts/titles: [insert]
1) Cluster them by question logic (institutional, actor-centered, intellectual history, political economy, gender).
2) Suggest 2 follow-up keywords per cluster (DE/FR/EN) and name 1 typical "blind spot" each.

Deliverable: Mini-cluster (5 items) + keyword update list.

5. Snowballing: Backward/Forward Citation (Control Canons and Blind Spots)

Goal

From a small core corpus, reconstruct footnotes/bibliographies in depth; simultaneously make canon bias visible.

Task (Without AI)

Select 3 “seed” texts (preferably different types: overview, monograph, article).

For each seed:

• 10 relevant references from bibliography/footnotes (backward),
• 5 newer works via citation search (forward; e.g., Google Scholar/publisher pages).

Document:

• why each reference was included (1 sentence),
• whether you expanded or narrowed a debate axis.

Task (AI Exploration)

Here are 10 references from the bibliography of a seed text: [insert]
Organize them by:
(a) debate axis, (b) method, (c) expected source base.
Mark: which 3 would be indispensable for a "State of Research" section – and justify.

Critical Check

Maintain a Reject Log: at least 5 rejected but “temptingly plausible” hits incl. rejection reason (wrong time period, wrong arena, only popular science, etc.).

Deliverable: Snowball list (≥ 30 references total) + Reject Log (≥ 5).

6. Screening and Quality Assessment: Relevance, Argument, Method, Evidence

Goal

Not just collect literature, but analytically select.

Task (Without AI)

Create an assessment rubric (1 page), e.g.:

• Research question/thesis
• Theory/concepts (explicit/implicit)
• Method (archive-based, discourse-analytical, comparative, quantitative)
• Source base (which primary sources? which gaps?)
• Contribution (what is new? what is contested?)
• Connection to your focus

Apply it to 6 titles from your corpus (short excerpt, 150–250 words per title).

Task (AI Exploration)

Use this rubric: [insert]
Create for each of the following titles/abstracts a structured brief annotation (max. 180 words),
incl. (1) presumed contribution, (2) methodological approach, (3) potential bias/limitation.
If information is missing: explicitly mark as "unclear," do not supplement.

Critical Check

Sample: check for 2 titles based on introduction/conclusion whether AI structure and your reading are consistent.

Deliverable: 6 annotated entries + rubric.

7. Historiographic Debate Mapping and State of Research Text (Condensed, Evidence-Sensitive)

Goal

Make the corpus a reconstructed “conversation”: Who argues how – and why?

Task (Without AI)

Write a 400–600 word debate memo:

• 2–3 main controversies,
• central positions,
• typical evidence or method disputes,
• open flanks (unaddressed, underexposed, contested).

Task (AI Exploration)

Here is my debate memo: [insert]
1) Identify implicit assumptions (agency/structure, state/private, normativity, contemporary diagnoses).
2) Suggest 2 alternative structurings (e.g., by arenas instead of topics).
3) Name 5 targeted follow-up questions I should ask the literature to formulate research gaps precisely.

Deliverable: Debate memo + 5 follow-up questions.

8. Research Gap and Delimitation: From State of Research to Research Question

Goal

Derive a workable, analytical research question from the literature review (not from “interest” alone).

Task (Without AI)

Formulate:

• 1 research gap statement (max. 3 sentences): “Research has shown X; Y remains unclear because…”
• 2 research questions (1 sentence each), which are plausible based on sources and literature.

Task (AI Exploration)

Based on this research gap: [insert]
Formulate 4 variants of an analytical historical research question.
For each variant:
- possible case delimitation (time/space/arena),
- expected contribution (which debate is addressed?),
- risk field (e.g., source access, anachronism, too broad causality).

Deliverable: 1 research gap + 1 prioritized research question + 1 risk section (150–200 words).

9. Reproducibility and Documentation

Goal

Literature work as a documented method: traceable, verifiable, reusable.

Minimum Requirements

• Search protocol: Date, system, query, filter, hits, selection rules (≥ 12 entries)
• Zotero/BibTeX export: Versioned state (date), duplicate strategy
• Decision log: Why title included/rejected (keywords)
• AI protocol (Appendix): 2–4 prompts, shortened responses, your corrections/rejects, purpose of use

WarningAI as Tool, Not as Source

AI may help in this exercise to expand search spaces and structure texts. It does not replace:

• bibliographic verification (catalog/DOI),
• source-critical examination of the evidence base,
• historiographic judgment formation.

Submission (Compact)

1. Focus statement + debate map (1–2 pages)
2. Search protocol (≥ 12 entries) + Reject Log (≥ 5)
3. Briefly annotated bibliography (10–15 titles)
4. Debate memo (400–600 words)
5. Research gap + research question + risk field (approx. 1 page)
6. AI protocol (Appendix)

Optional: Advanced Track (Digital Methods)

• Export metadata (title/author/year/keywords) from swisscovery/Zotero.

• Build a simple bibliography map:

  • Time course (publication years),
  • Languages/publication types,
  • Co-citation or keyword network (caution: catalog keywords are curated and biased).

• Document workflow and data (README + license/reuse notes).

Further Resources

• Ranke.2: Lessons Overview – Learning modules on web, social media, and digital change.
• Knowing Machines: Models All the Way Down – Critique of training data and its parallels to source criticism.

Bibliography

Milligan, Ian. 2019. History in the Age of Abundance? How the Web Is Transforming Historical Research. Montreal; Kingston: McGill-Queen’s University Press. https://doi.org/10.2307/j.ctvggx2kh.

Putnam, Lara. 2016. “The Transnational and the Text-Searchable: Digitized Sources and the Shadows They Cast.” The American Historical Review 121 (2): 377–402. https://doi.org/10.1093/ahr/121.2.377.

WarningAutomated Translation Disclaimer

This exercise was automatically translated from German using AI and may contain errors or inaccuracies. Please refer to the original German version for the authoritative text. If you notice any translation issues, please report them.