Pith. sign in

REVIEW 2 major objections 4 minor 72 references

Collective Memory and Narrative Cohesion: A Computational Study of Palestinian Refugee Oral Histories in Lebanon

T0 review · 2 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

Pith's one-line read In hundreds of Palestinian refugee oral histories from Lebanon, sharing a place of origin predicts how similarly people narrate the 1948 Nakba—the mass displacement of Palestinians—more than gender does.

desk verdict A serious new quantitative study of collective memory in POHA, but the semantic-embedding and transcript-NER results need an interviewer-speaker control, and the pair-count table has a mislabeled row, before the specific estimates can be trusted. read the letter →

arxiv 2501.13682 v1 pith:F3QGVYHE submitted 2025-01-23 cs.CL

classification cs.CL
keywords collectivememoryPalestinianNakbaoralhistorynarrativecohesionsemanticembeddingsmixed-effectsmodelsrefugeestudiescomputationalhumanities
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper asks whether Palestinian refugees in Lebanon, living in exile for generations, still remember the Nakba—the 1948 mass displacement—as a cohesive group rather than as isolated individuals. It answers by turning a large archive of transcribed oral-history interviews into vectors: one set counts the keywords, landmarks, families, and significant figures associated with each interview, and another uses language-model embeddings of transcript sections to capture broader semantic content. Pairwise cosine similarities between interviews become the outcome in statistical models (mixed-effects regressions) that test whether pairs sharing gender, village of origin, or current residence in Lebanon are more alike. The central finding is that shared origin is the strongest and most consistent predictor of narrative similarity across all representations; shared residence also predicts similarity and amplifies cohesion when combined with shared origin, while shared gender predicts only modest, context-specific cohesion, mainly in women's accounts of the British Mandate period. If true, this shows that collective memory is not just a metaphor: it leaves a measurable signal in how different narrators tell the same history.

What carries the argument

The machinery is a pairwise-similarity design applied to oral-history transcripts. Each interview is represented multiple ways: bag-of-words vectors built from archivist-curated metadata (thematic keywords, landmarks, families, significant figures), bag-of-words vectors from named entities extracted out of the Arabic transcripts, and dense semantic embeddings of transcript sections that have been classified into themes such as Zionist attacks, exile and expulsion, British Mandate colonialism, and resistance. For every pair of interviews, cosine similarity between the two vectors is computed, and that similarity score becomes the dependent variable in mixed-effects linear regressions with random intercepts for the two interviews, letting each interview have its own baseline while estimating how much pair-level sharing shifts similarity. The key predictors are binary indicators for whether the pair shares gender, shares place of origin, or shares place of residence, plus an interaction term for sharing both origin and residence; inverse-frequency weighting compensates for the rarity of pairs that share both, and Bayesian versions of the same models are used to check the frequentist intervals.

What would settle it

Using the speaker labels already present in the transcripts, remove all interviewer turns and recompute the pairwise cosine similarities and the mixed-effects coefficients; if the shared-origin and shared-residence effects shrink to nothing when only interviewee speech is embedded, the central claim is falsified. A complementary check is to add interviewer identity as a random effect to the models and see whether the geographic and gender coefficients survive.

Watch

Extended reading notes

Core claim

The paper's central claim is that collective memory of the Nakba survives in exile and is organized by geographic group boundaries. Interviewees who share the same place of origin produce more similar narratives than those with different origins—whether similarity is measured through archivists' keywords, mentions of landmarks, families, and significant figures, named entities extracted from transcripts, or semantic embeddings of whole thematic sections. Sharing a current residence in a Lebanese camp or locality also raises similarity, and interviewees who share both origin and residence show the strongest and most consistent cohesion, especially in themes tied directly to the Nakba: Zionist attacks, exile and expulsion, British Mandate colonialism, and resistance. Shared gender is a weaker, less uniform signal: women's narratives are noticeably more cohesive than men's in thematic content, particularly when recounting the British Mandate period and pre-Nakba village life, and this is interpreted as evidence that women's memories are shaped by communal roles and village life, not by gender alone. The authors offer this as large-scale empirical support for the idea that group memory operates through continuous spatial bonds to the homeland and through new bonds formed in exile.

Load-bearing premise

The load-bearing premise is that the interviewers' own words inside the transcripts do not distort the similarity measurements: the authors kept interviewer speech in the embeddings on the assumption that interviewers follow a highly consistent approach, so if interviewers differ by camp, village, or interviewee gender—asking different questions or talking in different ways—the apparent cohesion of the refugees' memories could partly be a mirror of the interviewers' behavior rather than of shared memory.

Editorial extensions

If this is right

  • Palestinian refugees from the same village or town in pre-1948 Palestine tell measurably more similar Nakba narratives than refugees from different origins, even when they now live in different places.
  • Sharing a current residence in Lebanon also raises narrative similarity, and refugees who share both origin and residence show the strongest cohesion, indicating that exile itself creates new memory frames.
  • Women's narratives are more thematically cohesive than men's in specific settings—notably the British Mandate period—while men's accounts are not similarly similar to each other.
  • The same pattern appears in both archivist-curated metadata and transcript embeddings, so the result is triangulated across two independent measurement approaches.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Beyond the paper: if interviewer speech is removed from the transcripts and the same-origin effects persist, the geographic-cohesion claim would be strengthened; if they shrink, the thematic-embedding results may partly measure interview protocol rather than shared memory.
  • Beyond the paper: a natural extension is to test whether mixed-origin camps show weaker shared-residence cohesion than camps where residents share a single origin, which would indicate that camp institutions transmit memory mainly within village networks.
  • Beyond the paper: the same pairwise-similarity plus mixed-effects design could be applied to other diaspora oral-history collections to ask whether origin-based memory cohesion generalizes across host countries or is specific to Lebanon's conditions of exclusion.
  • Beyond the paper: because the authors note that the archive's metadata reflects curatorial decisions, a transcript-only replication using named entities and embeddings would separate the refugees' own words from the archivists' descriptions.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 4 minor

Summary. The paper examines whether Palestinian refugee narrators in the POHA collection show measurable narrative cohesion along gender and geographic group boundaries. Using 724 interviews, the authors construct pairwise cosine-similarity measures across several textual representations: archivist-curated metadata bag-of-words vectors (keywords, significant figures, families, landmarks), transcript-derived named-entity bag-of-words vectors, and instruction-conditioned semantic embeddings of Table-of-Contents section excerpts. They fit mixed-effects and Bayesian models with dummy variables for gender pairings and for shared origin, shared residence, and their interaction. The main reported findings are that shared origin strongly predicts narrative similarity, shared residence also predicts similarity, sharing both is often a stronger predictor, and women's narratives are somewhat more cohesive, especially in themes such as the British Mandate and Zionist attacks. The results are interpreted through Halbwachs' theory of collective memory.

Significance. The contribution is potentially valuable as one of the first large-scale quantitative analyses of a Palestinian oral history archive with theory-driven group comparisons. The multi-representation design, using both metadata-based and transcript-based features, is a strength, as is the combination of frequentist and Bayesian mixed-effects models and the explicit acknowledgment that archival curation and metadata creation are sources of bias. If the findings survive the confounds identified below, they would provide empirical evidence about how shared geography and gender shape collective memory in diasporic oral histories and would offer a reusable computational template for oral-history research. The manuscript is transparent about many limitations, but the transcript-embedding prong requires an additional robustness check before the abstract's semantic-embedding claim can be accepted.

major comments (2)
  1. [4.2.2 / 4.3.2] In the semantic-embedding prong, embeddings are computed over unfiltered transcripts that include interviewer speech, and the only justification is a 'high degree of consistency' in interviewers' approach (Section 4.2.2). Consistency of interview structure does not imply consistency of lexical or stylistic choices, and the POHA metadata described in Section 4.1 includes interviewer identities that never enter Eq. (3) or the corresponding Bayesian models. If interviewer assignment correlates with camp, village, or interviewee gender, cosine similarities between same-location pairs will be inflated by shared interviewer wording and by interviewer mentions of the interviewee's origin or residence, independently of any collective memory among interviewees. The NER bag-of-words features are also computed over whole transcripts and are subject to the same mechanism. Because the abstract separately claims cohesion 'in semantic embeddings,' this is a load-bearing point. I ask for a speaker-role-filtered replication, or, at minimum, a model that includes interviewer identity as a random effect, and a discussion of whether the results persist under that specification.
  2. [Appendix B.4.1, Table 3] Table 3 is internally inconsistent with the surrounding text. The text states that 'less than 1% of the data involved pairs sharing both place of residence and place of origin,' but the table reports 251,036 pairs for 'Same Origin + Same Residence' and 634 pairs for 'Diff Origin + Diff Residence' among what must be roughly 274,000 total pairs. Either the row labels are swapped, so that the rarest cell is same-origin plus same-residence and the largest is diff-origin plus diff-residence, or the prevalence claim is incorrect. Since Section 4.3.2 says the inverse-frequency weights are based on the prevalence of each pairing type, and since the interaction term in Eq. (3) is the paper's key evidence for amplification, this inconsistency directly affects the validity of the location results in Figures 4 and 5. The table and text need to be reconciled and the weighted estimates recomputed or re-reported if the labels were wrong.
minor comments (4)
  1. [Eq. (1)] The notation in Eq. (1) uses σ(eA_j, eA_j) on both sides of the definition; the arguments should be eA_i and eA_j.
  2. [Appendix B.3 vs. Section 4.2.2] The main text says the embeddings were generated with OpenAI's text-embedding-3-large model, while Appendix B.3 names text-embedding-ada-002; this discrepancy should be corrected because it affects reproducibility.
  3. [Introduction / Section 6] There are several typos, including 'maintainance' in the Introduction and 'ohare' instead of 'share' in Section 6; a proofreading pass would improve readability.
  4. [Reproducibility] The appendices contain a detailed pipeline, but no code repository or data-access statement is mentioned; a short reproducibility statement would strengthen the paper.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper is an empirical association study whose statistical models do not assume the results they report.

full rationale

This is an empirical association study, not a derivation chain. The central claims are that shared origin and shared residence are associated with higher pairwise narrative similarity, measured through metadata BoW representations and transcript embeddings. The dependent variable in every model is the pairwise similarity score, and the independent variables are binary indicators of shared gender, origin, and residence; the coefficients are estimated rather than imposed, so the models do not assume the conclusions. No fitted parameter is renamed as a prediction, and no quantity is predicted from a subset of data that determines it by construction. The semantic-embedding analysis deliberately retains interviewer speech, justified by an external citation to Sleiman and Chebaro (2018); this is an archival description, not a self-citation by the authors, and the assumption is a construct-validity concern rather than a circular step. The shared-origin/landmark association is conceptually close to definitional because interviewees from the same village naturally mention the same village landmarks, but the paper treats this as a measurement choice and does not present it as a derived law, so it does not constitute circular reasoning. There is a minor internal inconsistency in the embedding model name between Section 4.2.2 and Appendix B.3, and there are legitimate validity risks around interviewer speech and metadata curation, but these are correctness or robustness issues, not circularity. No uniqueness theorem is imported, no ansatz is smuggled via self-citation, and no known result is merely renamed. The paper is self-contained against its empirical benchmarks, so the appropriate circularity score is 0.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

The central claim rests on a chain of operationalization choices: Halbwachs' theory is assumed, similarity is defined as cosine distance, metadata and transcriptions are taken as faithful, and interviewer speech is folded into embeddings. None of these are tested against an external benchmark, and the inverse weighting in the location models is affected by an apparent label swap in Table 3.

free parameters (3)
  • Inverse frequency weights for pair types = computed from pair counts (Table 3)
    The weighted location models depend on the inverse prevalence of each origin/residence pairing; the table's row labels appear inconsistent with the text, so the weights may be misassigned.
  • Bayesian priors = Normal(0,1) for β, Exponential(1) for σ
    Weakly informative priors chosen by hand in Appendix B.5; they influence posterior intervals but not the main qualitative findings.
  • SVC theme classifier = AUC 0.98 on test split
    Theme labels for TOC headers are produced by a one-vs-rest SVM trained on 340 manually annotated headers; errors propagate into theme-similarity analyses.
assumptions (6)
  • domain assumption Halbwachs' theory of collective memory applies to Palestinian refugees in Lebanon and predicts measurable narrative cohesion.
    The theoretical framework in Section 3 is assumed true and is the lens for interpreting all results.
  • domain assumption Cosine similarity between BoW and embedding representations is a valid operationalization of narrative cohesion.
    Section 4.2.3 defines cohesion via cosine similarity without external validation against qualitative judgments.
  • domain assumption Including interviewer speech in transcript embeddings does not confound group comparisons because POHA interviewers follow a consistent structure.
    Section 4.2.2 explicitly states interviewer speech was retained; no interviewer-identity covariate is included.
  • domain assumption Transkriptor transcriptions are accurate enough that transcription error does not systematically differ by origin, residence, or gender.
    Appendix A.5 reports a qualitative 90% accuracy assessment by two team members, with no quantitative error analysis by group.
  • domain assumption GPT-4o-extracted gender, origin, and residence fields are correct and complete.
    Appendices A.2 and A.3 describe LLM extraction and manual reading of about 250 transcripts, but no validation statistics are reported.
  • standard math Mixed-effects models with random intercepts for interviewees adequately handle the non-independence of pairwise similarity observations.
    Equations 2 and 3 assume normal random effects; this is a standard modeling assumption but unverified for this dependence structure.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Collective Memory and Narrative Cohesion: A Computational Study of Palestinian Refugee Oral Histories in Lebanon." pith.science (2026). https://pith.science/paper/F3QGVYHE

@misc{pith2026250113682,
  author       = {Pith},
  title        = {Pith review of: Collective Memory and Narrative Cohesion: A Computational Study of Palestinian Refugee Oral Histories in Lebanon},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/F3QGVYHE}},
  note         = {Machine review of arXiv:2501.13682}
}
read the original abstract

This study uses the Palestinian Oral History Archive (POHA) to investigate how Palestinian refugee groups in Lebanon sustain a cohesive collective memory of the Nakba through shared narratives. Grounded in Halbwachs' theory of group memory, we employ statistical analysis of pairwise similarity of narratives, focusing on the influence of shared gender and location. We use textual representation and semantic embeddings of narratives to represent the interviews themselves. Our analysis demonstrates that shared origin is a powerful determinant of narrative similarity across thematic keywords, landmarks, and significant figures, as well as in semantic embeddings of the narratives. Meanwhile, shared residence fosters cohesion, with its impact significantly amplified when paired with shared origin. Additionally, women's narratives exhibit heightened thematic cohesion, particularly in recounting experiences of the British occupation, underscoring the gendered dimensions of memory formation. This research deepens the understanding of collective memory in diasporic settings, emphasizing the critical role of oral histories in safeguarding Palestinian identity and resisting erasure.

Figures

Figures reproduced from arXiv: 2501.13682 by the authors.

Figure 1
Figure 1. UMAP of instruction-based embeddings of interview transcript sections for all themes. Related themes are visually indistinguishable in this decomposi￾tion, while different themes are distant from each other. Despite using the same measure for similarity, We heavily lean into the meanings of each of the aspects to interpret the meaning of similarity and cohesion contingent on what is being represented. 4.3 Statistica… view at source ↗
Figure 2
Figure 2. Model coefficients reflecting the relationship [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Model coefficients reflecting the relationship [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (7 more)
Figure 5
Figure 5. Figure 5: Model coefficients reflecting the relationship [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Mixed model for location estimates for all [PITH_FULL_IMAGE:figures/full_fig_p017_6.png]
Figure 7
Figure 7. Figure 7: Mixed model for gender estimates for all [PITH_FULL_IMAGE:figures/full_fig_p017_7.png]
Figure 8
Figure 8. Figure 8: Bayesian model estimates with credible in [PITH_FULL_IMAGE:figures/full_fig_p018_8.png]
Figure 9
Figure 9. Figure 9: Bayesian model estimates with credible in [PITH_FULL_IMAGE:figures/full_fig_p018_9.png]
Figure 10
Figure 10. Figure 10: Average cosine similarity between the embeddings of two interview segments on the labeled theme for [PITH_FULL_IMAGE:figures/full_fig_p019_10.png]
Figure 11
Figure 11. Figure 11: Cooccurrence between themes in POHA interview segments (i.e., the number of headers for which one [PITH_FULL_IMAGE:figures/full_fig_p020_11.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

72 extracted references · 62 canonical work pages

  1. [1]

    Salm \=a n Ab \=u -Sitta . 2016. Mapping My Return: A Palestinian Memoir . American University in Cairo Press, Cairo New York

  2. [2]

    Mousumi Akter and Shubhra Kanti Karmaker Santu. 2024. https://doi.org/10.48550/arXiv.2309.04823 FaNS : A Facet-based Narrative Similarity Metric . Preprint, arXiv:2309.04823

  3. [3]

    Amirah Allan. 2021. Introduction: Past Continuous . In Diana Allan, editor, Voices of the Nakba : A Living History of Palestine , pages 61--81. Pluto Press

  4. [4]

    Diana Allan. 2013. Refugees of the Revolution : Experiences of Palestinian Exile . Stanford University Press

  5. [5]

    David Bamman and Noah A. Smith. 2014. https://doi.org/10.1162/tacl_a_00189 Unsupervised Discovery of Biographical Structure from Text . Transactions of the Association for Computational Linguistics, 2:363--376

  6. [6]

    Bassam Yousef Ibrahim Banat, Francisco Entrena-Dur \'a n , and Jawad Dayyeh. 2018. https://doi.org/10.5539/ass.v14n12p147 Palestinian Refugee Youth : Reproduction of Collective Memory of the Nakba . ASS, 14(12):147

  7. [7]

    Rana Barakat. 2019. https://doi.org/10.1080/20581831.2019.1594617 Reading Palestinian agency in mandate history: The narrative of the Buraq Revolt as anti-relational . Contemporary Levant, 4(1):28--38

  8. [8]

    Efrat Ben-Ze'ev . 2002. https://doi.org/10.1080/02757200290002860 The Palestinian village of Ijzim during the 1948 war: Forming an anthropological history through villagers accounts and army documents . History and Anthropology, 13(1):13--30

Show all 72 references
  1. [9]

    Morris, Richard Antonello, Ion Stoica, Alexander G

    Vinamra Benara, Chandan Singh, John X. Morris, Richard Antonello, Ion Stoica, Alexander G. Huth, and Jianfeng Gao. 2024. https://doi.org/10.48550/ARXIV.2405.16714 Crafting Interpretable Embeddings by Asking LLMs Questions

  2. [10]

    Filip Bendevski, Jumana Ibrahim, Tina Krulec, Theodore Waters, Nizar Habash, Hanan Salam, Himadri Mukherjee, and Christin Camia. 2021. https://doi.org/10.1145/3462244.3479895 Towards Automatic Narrative Coherence Prediction . In Proceedings of the 2021 International Conference...

  3. [11]

    a ge zum ausl \

    Eyal Benvenisti, Chaim Gans, and Sari Hanafi, editors. 2007. https://doi.org/10.1007/978-3-540-68161-8 Israel and the Palestinian Refugees . Number 189 in Beitr \"a ge zum ausl \"a ndischen \"o ffentlichen Recht und V \"o lkerrecht . Springer Berlin Heidelberg, Berlin, Heidelberg

  4. [12]

    Bisharat

    George E. Bisharat. 1994. https://doi.org/10.1111/j.2050-411X.1994.tb00806.x Displacement and Social Identity : Palestinian Refugees in the West Bank . Cent Migr Stud Spec Iss, 11(4):163--188

  5. [13]

    Patrick Brookshire and Nils Reiter. 2024. https://aclanthology.org/2024.latechclfl-1.10 Modeling Moravian Memoirs: Ternary Sentiment Analysis in a Low Resource Setting . In Proceedings of the 8th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social ...

  6. [14]

    Rogers Brubaker. 2005. https://doi.org/10.1080/0141987042000289997 The `diaspora' diaspora . Ethnic and Racial Studies, 28(1):1--19

  7. [15]

    Paul-Christian B \"u rkner. 2017. https://doi.org/10.18637/jss.v080.i01 Brms: An R Package for Bayesian Multilevel Models Using Stan . Journal of Statistical Software, 80:1--28

  8. [16]

    Kim D. Butler. 2001. https://doi.org/10.1353/dsp.2011.0014 Defining Diaspora , Refining a Discourse . dsp, 10(2):189--219

  9. [17]

    Carbajal and Michelle Caswell

    Itza A. Carbajal and Michelle Caswell. 2021. https://doi.org/10.1093/ahr/rhab359 Critical Digital Archives : A Review from Archival Studies . The American Historical Review, 126(3):1102--1120

  10. [18]

    Giovanni Colavizza, Tobias Blanke, Charles Jeurgens, and Julia Noordegraaf. 2021. https://doi.org/10.1145/3479010 Archives and AI: An Overview of Current Debates and Future Perspectives . Journal on Computing and Cultural Heritage, 15(1):4:1--4:15

  11. [19]

    Rochelle A. Davis. 2011. Palestinian Village Histories: Geographies of the Displaced. Stanford Studies in Middle Eastern and Islamic Societies and Cultures. Stanford University Press, Stanford, Calif

  12. [20]

    Farah Aboubakr . 2017. https://doi.org/10.13110/marvelstales.31.2.0217 Peasantry in Palestinian Folktales : Sites of Memory , Homeland , and Collectivity . Marvels & Tales, 31(2):217

  13. [21]

    Edwin Farley and R. Gutman. 2020. A Bayesian Approach to Linking Data Without Unique Identifiers . arXiv: Computation

  14. [22]

    Jane Greenberg. 1998. https://doi.org/10.17723/aarc.61.2.j3p8200745pj34v6 The Applicability of Natural Language Processing (NLP) to Archival Properties and Objectives . The American Archivist, 61(2):400–425

  15. [23]

    Maurice Halbwachs. 1992. On Collective Memory . University of Chicago Press

  16. [24]

    Knudsen, editors

    Sari Hanafi and Are J. Knudsen, editors. 2011. https://doi.org/10.4324/9780203839256 Palestinian Refugees: Identity, Space and Place in the Levant . Routeldge, Milton Park, Abingdon, Oxon New York

  17. [25]

    Isabelle Humphries and Laleh Khalili. 2007. Gender of Nakba Memory . In Ahmad H. Sa'di and Lila Abu-Lughod , editors, Nakba: Palestine , 1948, and the Claims of Memory , Cultures of History, pages 207--228. Columbia University Press, New York

  18. [26]

    Go Inoue, Bashar Alhafni, Nurpeiis Baimukan, Houda Bouamor, and Nizar Habash. 2021. The Interplay of Variant, Size, and Task Type in A rabic Pre-trained Language Models . In Proceedings of the Sixth Arabic Natural Language Processing Workshop, Kyiv, Ukraine (Online). Associati...

  19. [27]

    Lise Jaillant and Arran Rees. 2023. https://doi.org/10.1093/llc/fqac073 Applying AI to digital archives: trust, collaboration and shared professional ethics . Digital Scholarship in the Humanities, 38(2):571–585

  20. [28]

    Lena Jayyusi. 2007. Iterability, Cumulativity , and Presence : The Relational Figures of Palestinian Memory . In Ahmed H. Sa'di and Lila Abu-Lughod , editors, Nakba: Palestine , 1948, and the Claims of Memory , Cultures of History, pages 107--134. Columbia University Press, New York

  21. [29]

    Rashid Khalidi. 1997. Palestinian Identity . Columbia University Press

  22. [30]

    Walid Khalidi. 1992. All That Remains : The Palestinian Villages Occupied and Depopulated by Israel in 1948 . Institute for Palestine Studies

  23. [31]

    Elias Khoury. 2012. https://doi.org/10.1086/662741 Rethinking the Nakba . Critical Inquiry, 38(2):250--266

  24. [32]

    Laura Khoury. 2018. Shu'fat refugee camp women authenticate an old `` Nakba '' and frame something ``new'' while narrating it. In Nahla Abdo-Zubi and Nur Masalha, editors, An Oral History of the Palestinian Nakba , pages 136--158. ZED, London, UK

  25. [33]

    Owen C. King. 2024. https://doi.org/10.1007/s10502-024-09443-z Archival meta-metadata: Revision history and positionality of finding aids . Arch Sci, 24(3):509--529

  26. [34]

    Nur Masalha. 2012. The Palestine Nakba : Decolonising History, Narrating the Subaltern, Reclaiming Memory . Zed Books, London

  27. [35]

    Nur Masalha. 2018. Decolonizing methodology, reclaiming memory: Palestinian oral histories and memories of the Nakba . In Nahla Abdo-Zubi and Nur Masalha, editors, An Oral History of the Palestinian Nakba , pages 6--39. ZED, London, UK

  28. [36]

    Joseph Andoni Massad. 2006. The Persistence of the Palestinian Question: Essays on Zionism and the Palestinians . Routledge Kegan Paul, London New York, NY

  29. [37]

    Leland McInnes, John Healy, Nathaniel Saul, and Lukas Großberger. 2018. https://doi.org/10.21105/joss.00861 UMAP: Uniform Manifold Approximation and Projection . Journal of Open Source Software, 3(29):861

  30. [38]

    Roxy Moon. 2023. https://doi.org/10.1080/0377919X.2024.2304322 Outside the Locus of Control : Palestinian Digital Archives Resist Israeli Settler-Colonial Erasure . Journal of Palestine Studies, 52(4):37--52

  31. [39]

    Sonia Nimr. 2008. https://doi.org/10.4000/clo.287 Fast Forward to the Past : A Look into Palestinian Collective Memory . clo, 63--64:338--349

  32. [40]

    Ossama Obeid, Nasser Zalmout, Salam Khalifa, Dima Taji, Mai Oudah, Bashar Alhafni, Go Inoue, Fadhl Eryani, Alexander Erdmann, and Nizar Habash. 2020. https://www.aclweb.org/anthology/2020.lrec-1.868 CAM e L Tools: An Open Source Python Toolkit for A rabic Natural Language Proc...

  33. [41]

    https://openai.com/index/hello-gpt-4o/ Hello GPT-4o

    OpenAI. https://openai.com/index/hello-gpt-4o/ Hello GPT-4o

  34. [42]

    OpenAI, Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, Red Avila, Igor Babuschkin, Suchir Balaji, Valerie Balcom, Paul Baltescu, Haiming Bao, Mohammad Bavarian, Jeff ...

  35. [43]

    Ilan Papp \'e . 2007. The Ethnic Cleansing of Palestine . Oneworld Publications, New York

  36. [44]

    Francisca Pessanha and Almila Akdag Salah. 2021. https://doi.org/10.1145/3477605 A Computational Look at Oral History Archives . Journal on Computing and Cultural Heritage, 15(1):6:1--6:16

  37. [45]

    Andrew Piper. 2023. https://doi.org/10.18653/v1/2023.bigpicture-1.3 Computational Narrative Understanding: A Big Picture Analysis . In Proceedings of the Big Picture Workshop, pages 28--39, Singapore. Association for Computational Linguistics

  38. [46]

    Mazin B Qumsiyeh. 2011. Popular Resistance in Palestine : A History of Hope and Empowerment . Pluto Press, New York, NY

  39. [47]

    Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, and Ilya Sutskever. 2022. https://arxiv.org/abs/2212.04356 Robust Speech Recognition via Large-Scale Weak Supervision . Preprint, arXiv:2212.04356

  40. [48]

    https://beautiful-soup-4.readthedocs.io/en/latest/index.html Beautiful soup

    Leonard Richardson. https://beautiful-soup-4.readthedocs.io/en/latest/index.html Beautiful soup

  41. [49]

    sumud is to always be one hand

    Rami Rmeileh. 2021. “sumud is to always be one hand”: Culturally informed resilience among palestinian refugee men in lebanon. Master's thesis

  42. [50]

    Sa'di and Lila Abu-Lughod

    Ahmad H. Sa'di and Lila Abu-Lughod . 2007. Introduction: The Claims of Memory . In Ahmed H. Sa'di and Lila Abu-Lughod , editors, Nakba: Palestine , 1948, and the Claims of Memory , Cultures of History, pages 1--24. Columbia University Press, New York

  43. [51]

    William Safran. 1991. https://doi.org/10.1353/dsp.1991.0004 Diasporas in Modern Societies : Myths of Homeland and Return . dsp, 1(1):83--99

  44. [52]

    Brenda Santana, Ricardo Campos, Evelin Amorim, Al \'i pio Jorge, Purifica c \ a o Silvano, and S \'e rgio Nunes. 2023. https://doi.org/10.1007/s10462-022-10338-7 A survey on narrative extraction from textual data . Artif Intell Rev, 56(8):8393--8435

  45. [53]

    Rosemary Sayigh. 1979. The Palestinians : From Peasants to Revolutionaries . Zed Books Ltd

  46. [54]

    Rosemary Sayigh. 1998. https://doi.org/10.2307/2538283 Palestinian Camp Women as Tellers of History . Journal of Palestine Studies, 27(2):42--58

  47. [55]

    Rosemary Sayigh. 2007. https://doi.org/10.2979/MEW.2007.3.1.86 Product and Producer of Palestinian History : Stereotypes of `` Self '' in Camp Women 's Life Stories . Journal of Middle East Women's Studies, 3(1):86--105

  48. [56]

    Rosemary Sayigh. 2015. https://doi.org/10.1080/2201473X.2014.955945 Oral history, colonialist dispossession, and the state: The Palestinian case . Settler Colonial Studies, 5(3):193--204

  49. [57]

    David Shuttleworth and Jose Padilla. 2022. https://doi.org/10.1109/WSC57314.2022.10015274 From Narratives to Conceptual Models via Natural Language Processing . In 2022 Winter Simulation Conference ( WSC ) , pages 2222--2233

  50. [58]

    Rami Siklawi. 2019. https://doi.org/10.13169/arabstudquar.41.1.0078 The Palestinian Refugee Camps in Lebanon Post 1990: Dilemmas of Survival and Return to Palestine . Arab Studies Quarterly, 41(1)

  51. [59]

    Amirah Silmi. 2021. The Margin and the Centre in Narrating Pre-1948 Palestine . In Diana Allan, editor, Voices of the Nakba : A Living History of Palestine , pages 61--81. Pluto Press

  52. [60]

    Hana Sleiman and Kaoukab Chebaro. 2018. https://doi.org/10.1525/jps.2018.47.2.63 Narrating Palestine : The Palestinian Oral History Archive Project . Journal of Palestine Studies, 47(2):63--76

  53. [61]

    Stan Development Team . 2024. RStan : The R interface to Stan

  54. [62]

    Smith, Luke Zettlemoyer, and Tao Yu

    Hongjin Su, Weijia Shi, Jungo Kasai, Yizhong Wang, Yushi Hu, Mari Ostendorf, Wen-tau Yih, Noah A. Smith, Luke Zettlemoyer, and Tao Yu. 2023. https://doi.org/10.18653/v1/2023.findings-acl.71 One Embedder, Any Task: Instruction-Finetuned Text Embeddings . In Findings of the Asso...

  55. [63]

    Ted Swedenburg. 1995. Memories of Revolt: The 1936 - 1939 Rebellion and the Palestinian National Past . University of Minnesota Press, Minneapolis, Minn

  56. [64]

    Szojka, Annabelle Nicol, and David La Rooy

    Zsofia A. Szojka, Annabelle Nicol, and David La Rooy. 2020. https://doi.org/10.1002/acp.3673 Narrative coherence in multiple forensic interviews with child witnesses alleging physical and sexual abuse . Applied Cognitive Psychology, 34(5):943--960

  57. [65]

    Salim Tamari. 2008. Mountain against the Sea : Essays on Palestinian Society and Culture . University of California Press

  58. [66]

    Mari Toivanen and Bahar Baser. 2019. https://doi.org/10.3138/gsi.13.1.02 Remembering the Past in Diasporic Spaces : Kurdish Reflections on Genocide Memorialization for Anfal . Genocide Studies International, 13(1):10--33

  59. [67]

    Transkriptor . n.d. Transkriptor API

  60. [68]

    Sofiya Voytiv. 2024. https://doi.org/10.1080/01419870.2023.2261289 Diasporic group boundaries and solidarity in the making: Collective memory in the anti-war protests in Sweden . Ethnic and Racial Studies, 47(2):391--410

  61. [69]

    Abbad Yahya. 2017. Oral History and Dual Marginalization :. Jerusalem Quarterly

  62. [70]

    Lixing Zhu, Runcong Zhao, Lin Gui, and Yulan He. 2023. https://doi.org/10.48550/arXiv.2310.18783 Are NLP Models Good at Tracing Thoughts : An Overview of Narrative Understanding . Preprint, arXiv:2310.18783

  63. [71]

    online" 'onlinestring :=

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint eprinttype howpublished institution journal key month note number organization pages publisher school series title type volume year doi pubmed url lastchecked label extra.label sort.label short.list...

  64. [72]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.