← Don't Scroll Lab
Venezuela exodus

Two parallel regional processes · Cartagena (refugee protection) and the South American Conference on Migration · 1984 to 2025

Cartagena and the South American Conference on Migration: two regional tracks

Two regional processes shaped how the Americas talk about forced and mixed movement. The Cartagena track deals with refugee protection in Latin America and the Caribbean, from the 1984 colloquium to the Chile Plan of Action 2024–2034. The CSM track is the South American Conference on Migration, founded in 2000 by the South American governments to coordinate migration policy. They ran separately, with different forums and different vocabulary. On this page they sit side by side, and a link is shown only where one text explicitly cites the other: in 2004, 2010–11, 2014–15, 2016–18, 2022 and 2024. The Quito Process, on its own tab, sits next to both. Analysis in English. Quotations are literal, in the language of the file supplied (Spanish or English), with references. Nothing is listed unless the text says it. The content was processed with artificial intelligence (Claude, by Anthropic); see Methodology.

Two tracks and where they meet

  • Cartagena track · refugee protection · 1984 to 2024Colloquia (1984, 1994), then decennial declarations and plans of action (2004, 2014, 2024), the Brasília Declaration (2010), situation-specific frameworks (San José 2016, MIRPS 2017), two global texts (New York 2016, Global Compact on Refugees 2018), the IACHR Principles (2019) and Los Ángeles (2022).
  • CSM track · migration governance · 2000 to 2025Annual conferences of the South American States with IOM as Technical Secretariat; two plans of action (2002, 2010), Rules (2016), six thematic Networks (2021) and the Plan 2035. The vocabulary is regularization, free movement and South American citizenship; refugees appear in preambles and in one text on disasters.
  • Where they meetFrom the CSM side, the last citation of the other track is in 2018 (the disaster Guidelines). From the Cartagena side, citations of the CSM come back in 2022 and 2024. See the tab “Where the tracks meet”.

The tracks, the digests and the notes on how they relate are an analytical grouping; the texts do not use these labels. Two Cartagena-track texts are global (New York 2016, Global Compact 2018) and one is a human-rights standard (IACHR 2019); they are included because the regional texts after them build on them.

Who the title names (Cartagena track)

Coded from each title only, not from the content. Los Ángeles names “Migración”, coded as migrants; the 2016 San José title names no group. The CSM texts are titled by place and year (“Declaración de Caracas”) or by motto, and name no group; they are not coded. Hover or focus a row for the full title.

Cartagena trackCSM trackDigest of several declarationsdashed chip a text that cites, or is cited by, a text of the other trackA blank lane in a year means no text of that track in the corpus that year.

Sources and coverage

What was read

Gaps and file issues

  • The declaration of the XXII CSM (2024) was not found and is not in the corpus. It is shown as a gap on the timeline. The only 2024 CSM text read is the Plan 2035, which records a plenary session on 6 November 2024 and asks the Technical Secretariat for a financing proposal.
  • The file named “2016_DeclaracaoAccionSanJose” is not the San José Action Statement. It is the two-page concept note for a side event at the UN Summit for Refugees and Migrants (21 September 2016) that presents the Statement. The card describes the Statement only as the note summarises it; the Statement’s own commitments are not in the corpus.
  • The file named “2021-CSM_-_declaracion_final” is the II CSM (Santiago, 2–3 April 2001), not a 2021 text; the XIX CSM (2021) is “2021CSM-Declaracio_n_ES”. The III CSM took place on 15–16 August 2002 according to the signed declaration. The Declaración de Principios file carries no date or place of its own; its approval at the X CSM comes from the Acta. The PSDHM’s cover says “aprobado” but its introduction calls it a discussion document, not to be cited without authorisation; it is read here as the plan approved on 25 October 2010 per the Acta.
  • Plan 2035: two editions are supplied (December 2024, 52 pages; June 2025, 55 pages with a list of contributors). The text gives no date of approval; the page uses the June 2025 edition and quotes from it. The two editions were compared: the December one carries a note 9 saying that “human mobility” refers “exclusively to its migratory dimension”, which the June one drops (it is quoted on the Plan 2035 card). The Plan dates the La Paz declaration to 2005 (p. 32), whereas the declaration is dated 25–26 November 2004, and it quotes the XIII CSM as “un migrante ilegal es un migrante invisible”, where the English text of that declaration says “irregular”.
  • The Mexico 2004, Brazil 2014 and Chile 2024 files are English versions. Mexico and Brazil say “Original: Spanish”, so their quotations are from the English translation, not the authentic Spanish text. The file named “2010_DeclaracaoBrasilia” is in Spanish. The Cartagena file holds Spanish, French and English versions; the Spanish one was used. On the CSM track, the 2002 Plan of Action, and the declarations of 2003, 2004, 2006, 2008, 2011, 2013 (XIII), 2014, 2015, 2016 and 2018 (Sucre), the Lima declaration of 2017, the IMRF statement of 2022, the Regional Guidelines of 2018 and the Lima declaration of 2023 were supplied in English; the others are in Spanish. English texts may differ from the Spanish originals, and some are uneven (the 2008 text speaks of the “Residence and Immigration Regularization of the MERCOSUR”).
  • The Mexico 2004, Brazil 2014 and Chile 2024 files do not list the governments that adopted them, so no count is given. The San Pedro Sula file prints no date; its year comes from the file name. For the CSM, the delegation lists in the files differ from year to year: the XIII CSM (2013) lists ten delegations and not Venezuela; the XVI (2016) signature block prints Colombia twice and not Argentina; the XIX (2021) lines of Guyana and Suriname are blank. Each card says what its text prints.
  • Six texts are scanned images without a text layer: the Los Ángeles Declaration (Cartagena track) and, on the CSM track, the IX (2009), XII (2012), XVII (2017), XIX (2021) and XXI (2023) declarations. They were read page by page from images by the AI model; — quotations were transcribed from them and not machine-checked, and are worth a spot check before quoting in print. For the XIX declaration only the declaration (pp. 1–5) was read, not the annexes with the Networks’ reports. The XX CSM declaration (2022) has a text layer made by OCR with some garbled words; quotations from it avoid them.
  • Two texts are global. In the New York Declaration only the parts on refugees, the comprehensive refugee response framework (Annex I) and the general commitments relevant to protection were coded; the commitments for migrants (section III) and Annex II were not. In the Global Compact on Refugees, the arrangements for responsibility-sharing, the protection and solutions sections and the follow-up were coded; the sector-by-sector areas of support were not. The IACHR Principles have 80 principles; those on definitions, non-refoulement, asylum access, regularization, documents, detention, statelessness, xenophobia and cooperation were coded.
  • The Regional Guidelines of 2018, the PSDHM and the Plan 2035 (June 2025 edition) were first read in part and then, in a second pass on 26 October 2026, read in full, including notes, annexes and lists of contributors; the December 2024 edition was compared with the June one. Page references to the Guidelines and the PSDHM are the printed page numbers. The Guidelines carry a leftover query mark in their text (“(?)”, p. 22) and cite an unpublished study (Cantor, 2018).
  • Texts cited by the corpus but not supplied, and so not described: the 1981 Mexico colloquium, CIREFCA (1989), the Brazil Plan’s triennial and final reports, the Brasilia 100 Points (2018), MIRPS annual reports, the Global Compact for Safe, Orderly and Regular Migration, the Palenque Declaration, the Brasilia Consensus, the CSM’s Declaration of Lima of 1999, the “Visión Regional para el Abordaje de las Migraciones en América del Sur” validated at the XXI CSM (2023) and described by the Plan 2035, and the Annexes of the XIX and XX declarations (the Networks’ reports, the COP27 joint position and the diaspora roadmap).
  • The texts contain no stable web links. Each file name links to the PDF copy supplied for this analysis, hosted with this page (docs/regional/ and docs/csm/). It is not an official URL.
  • The tracks, the digests, the one-line “shift” of each card, the theme labels, the “what the text does not say” notes, the dates of silences and the optional “link to the Quito Process” notes are the AI model’s analysis. Everything else restates the text. See Methodology.
Methodology

Transparency note. The information on this page was processed with artificial intelligence. An AI model (Claude, by Anthropic) read the documents, extracted and classified their content, wrote the summaries in English, drafted the text and built the page, working from PDF files supplied by the project owner, who set the scope, asked for a timeline in the style of the Quito Process page and then asked for the South American Conference on Migration to be added as a parallel track. Prepared on 25–26 October 2026. The output can contain errors of reading, classification or summary. Check it against the linked PDFs before quoting it.

Steps

  1. Scope. The corpus is — PDF files: twelve texts of the Cartagena track (the Cartagena Declaration 1984 and its commemorations San José 1994, Mexico 2004, Brazil 2014 and Chile 2024; the Brasília Declaration 2010; the concept note on the San José Action Statement 2016; the New York Declaration 2016; the San Pedro Sula Declaration 2017; the Global Compact on Refugees 2018; the IACHR Principles 2019; the Los Ángeles Declaration 2022) and thirty-one files of the CSM track: the final declarations and extraordinary declarations supplied for the Conferences of 2000 to 2023, the Plan of Action of 2002, the Declaración de Principios, the Acta and the PSDHM of 2010, the Rules of 2016, the Regional Guidelines of 2018 and two editions of the Plan 2035. No outside source was used for the content. The Quito Process card and the optional notes draw only on the Quito Process page of this site.
  2. Reading. Files with a text layer were extracted as text. Six scanned texts were read page by page from images; one more has an OCR layer (see the gaps list).
  3. Identification. Title, date, place and participants were taken from the text, not the file name. This showed that the 2016 San José file is a concept note, that the 2010 Brasília file is in Spanish, that “2021-CSM_-_declaracion_final” is the 2001 Conference, and that two files are two editions of the same Plan.
  4. Extraction. Each document was broken down into items: a statement, appeal, recommendation, programme, body or funding request that the text itself makes, with its paragraph, section or page reference. The rule was to record only what is explicitly written. Global texts and the IACHR Principles were coded selectively. For the CSM, the Regional Guidelines, the PSDHM and the Plan 2035 were read in full on a second pass and their items revised accordingly; the key declarations got a card each and eleven others are grouped in three digests (2001–2009, 2012–2013, 2017) that keep only the points relevant to this page. The result is — items across — cards.
  5. Classification. Each item was given one theme (—) and one type of intervention (6). Both lists were made for this page and do not come from the texts. The thirteen first themes are those of the refugee texts; five were added so that the content of the CSM texts has a place: Free movement of persons and Residence & citizenship (the project owner asked for two themes rather than one broad “free movement and integration” theme), and Migrants’ human rights & access to justice, Labour migration, nationals abroad & remittances and Information, data & training (added by the AI model; the project owner has not validated these three). A few existing items on the Cartagena track were re-coded to the new themes or split, and items were added on human rights to the New York, IACHR and Los Ángeles texts, so that the two tracks can be read against one list. The six types are the same as on the Quito Process page: Stance (a position, recognition or reaffirmation); Appeal (a call to States, the international community or other bodies); Recommendation to States (a measure States are invited to take); Instrument or product (a programme, study, handbook or guideline); Institution or mechanism (a body, platform, meeting cycle or follow-up arrangement); Funding call (a request for financial or technical cooperation).
  6. Quotations. The — quotations are literal, in the language of the file, with their reference. For the files with a text layer each quotation was checked automatically against the extracted text (quotations spanning a page break were checked ignoring page numbers). The quotations from scanned texts were transcribed from the page images and not machine-checked. Items in English are AI summaries, not official translations.
  7. Citations between the tracks. A link is listed only where one text names a text or process of the other track, and the passage is quoted. Overlaps of subject matter that the texts do not connect (for example regularization, children, disaster displacement) are not drawn as links; they can be seen in the matrix. “What the text does not say” notes come from a search of the text layers for the Quito Process, R4V, the Cartagena Declaration and the Mexico, Brazil and Chile plans, and from reading the scanned pages; they apply to the texts read, not to every document of the CSM.
  8. Matrix. A number is the count of items coded to that theme in that document. A dot means no item was coded, not that the topic is absent from the text. Counts depend on how sentences were split and on how selectively a text was coded, and the five added themes were built from the CSM’s vocabulary, so they score high for the CSM by construction. They show where a document puts its attention, not how important a theme is.
  9. Who the title names. Each title of the Cartagena-track texts was read for the groups it names (refugees, displaced, stateless, migrants, victims of trafficking). Only the title was used.
  10. Follow-up architecture. For each text, who adopts it and what follow-up body, cycle and reporting it sets up, as written in the text.
  11. Tracks, digests, shifts and the Quito link. The two tracks, the grouping of eleven CSM files in three digests, the one-line “shift” of each card and the optional notes linking each text to the Quito Process are an analytical reading made for this page. The texts do not use these labels.
  12. Page checks. The page was opened in a browser at desktop and phone width, in light and dark mode, to check that it renders, that the filters, tabs, track switch and hover texts work and that every PDF link points to an existing file. It was not tested on a physical phone or on a deployed preview.

Limits

  • The classification is a judgement. A different reader could split or label some items differently.
  • Many texts were read in English translation; wording in the Spanish originals may differ.
  • Only the supplied texts are covered. The CSM’s 2024 declaration is missing and the CSM’s other documents (technical reports, Networks’ annexes) are mostly outside. Whether a programme was applied, or a definition incorporated into a given country’s law, is outside this page.
  • Telling the CSM and Cartagena apart is itself a choice: governments take part in both, and a text can belong to more than one process. The page shows what each text says about the other.