docs/research/history-sint-maarten-gap-2026-10-23.md +102 −0
@@ -0,0 +1,102 @@
+# Sint Maarten: prosecution policy and legalisation proposals
+
+Researched 23 October 2026. This batch adds six history entries, revises the existing 2018
+prosecution entry, writes three articles and converts all twelve profile text blocks to inline
+citations. It extends Sint Maarten from two entries to eight, including the retained first-ban
+estimate. Tags cover the substance of each event; proposals remain in the debate category.
+
+## Sources read and dates
+
+The official Overheid.nl legislative metadata records the 28 March 1960 ordinance commencing
+on **1 June 1964** under the decree of 29 April 1964. The archived Netherlands Antilles
+consolidation expressly replaces the ordinance of 30 July 1929. This establishes continuity
+of drug controls, but does not recover the earliest cannabis-specific ban. The pre-existing
+estimated first restriction and its uncertainty window remain unchanged. Later consolidated
+cannabis definitions, exemptions and penalties are not backdated to 1960 or 1964. The current
+catalogue's 2015 version date alone does not establish a new penalty event.
+
+The original prosecution guideline PDF was read at pp. 1–2 and 5–6, with relevant surrounding
+text, from the retained Dutch Caribbean research files. It applies to Sint Maarten, Curaçao
+and the Caribbean Netherlands from **1 December 2018**, replacing July 2015 guidance. The
+0–25 g band provides confiscation and potentially conditional dismissal. The 25–50 g band
+provides a fine request up to NAf 500; from 50 g the cocaine/heroin requests are halved.
+The table overlaps at its printed boundaries, so entries retain its own bands. These are
+prosecutors' decisions and sentencing requests, not observed judicial outcomes. Exceptions
+include minors, repeat offending, dealing indications, another offence and specified settings.
+The separate discussion calls for prosecution of minors even with 0–5 g to support assistance.
+Grow cases receive individual assessments. Aruba's different thresholds are not imported.
+
+The full September 2022 government announcement was read in its NEWS.SX reproduction.
+It gives 12 September as launch and 5 December as deadline, despite also saying 90 days.
+The fixed deadline is used; no duration is inferred. The announcement invites studies and
+proposals, and does not document a completed consultant study or its findings.
+
+The full six-page **AB 2024 no. 11** regulation was read as web-extracted PDF text, including
+its explanatory notes. Its search metadata incorrectly names petroleum prices; its actual body
+is the cannabis committee regulation. It takes effect on signing, **28 February 2024**
+(the justice minister's signature is dated 27 February), and was published **6 March**.
+It creates an inter-ministerial committee and a programme of research and recommendations;
+it does not itself grant adult-use possession, cultivation or retail rights. Its description
+of Aruba is a comparison the committee was tasked to consider, not independent verification
+of Aruba's current medical access.
+
+Parliament's March notice and April follow-up were read in full. The latter confirms the
+13 March 2026 meeting was adjourned and announces a 9 April continuation. The March event
+is corroborated by the full SMN News report of Native Nations SXM's presentation. Industry
+job, tax and economic forecasts were not treated as established outcomes or included as facts.
+No later hearing date is inferred from inconsistent annual-report search snippets.
+
+The full Daily Herald report, syndicated by SXM Talks on 25 August 2026, describes remarks
+made during the Monday budget debate: **24 August** is the event date. The full relevant
+SMN News report of **14 October 2026** records Claret Connor seeking PAHO assistance.
+It establishes a request, not PAHO acceptance or a completed assessment. Ministers' statements
+about health risks are reported as their policy concerns; secondary clinical claims are not
+used as independently verified medical evidence.
+
+## Profile and score treatment
+
+The current official ordinance was read directly, particularly articles 1, 2a, 3a, 4 and 6–8.
+Article 7(3) expressly exempts possession of lawfully obtained amounts needed for personal
+medical use, alongside professional medical uses. Article 7(2) covers authorised pharmacists
+and practitioners. The profile's medical label changes from **none to limited** to reflect
+these statutory exemptions. An operating cannabis patient programme, pharmacy inventory or
+routine prescribing pathway was not verified. Unsupported travel-guide claims of a 2019
+programme are not repeated. The profile spells out the exemption and the ongoing review.
+
+All six numeric scores, dependency, recreational status, club status and visitor status are
+preserved. No proposal is scored as an implemented reform. No broad statutory maximum is
+presented as the normal punishment for a small personal quantity. The profile and articles
+explain the prosecution policy in terms of requests rather than automatic court sentences.
+
+## Retrieval limits and preserved evidence
+
+Local research inputs, extracted text, snapshots, source metadata, batch scripts and logs are
+in `tmp/history-sint-maarten-gap-2026-10-23/`. The earlier guideline PDF and inspected page
+images remain in `tmp/pdfs/dutch-caribbean-2026-10-20/`.
+The supposed `sint-maarten-2019.pdf` there is HTML, not a usable PDF, and was not cited as one.
+Direct retrieval of the 2024 government PDF failed certificate verification; TLS verification
+was not disabled. Its actual full text was read through the web cache. A screenshot request
+failed, so visual inspection of that PDF is not claimed. Direct SMN requests returned 403;
+the substantive article bodies were read through web retrieval. Industry workgroup pages
+were leads only, not evidence of enacted law.
+
+## Validation
+
+The scoped verifier checks all seven touched events, valid categories and tags, title lengths,
+checked status, related IDs, inline references and actual use of every listed source. The three
+articles have 388, 393 and 399 words and four headings each, with country and earlier/later
+history links. All twelve profile blocks pass key and source checks. The numeric profile vector
+matches the baseline. Unrelated concurrent edits in twelve existing events and six additions
+are preserved and recorded in `concurrent-revisions.json`; these include commencement-date
+corrections for Uganda, New Zealand and India and current-law reviews in other countries.
+
+The final full build passed: **3,824 pages in 30.02 seconds**. All **121 rendered citation
+links** on the seven touched history pages and country profile resolve to unique, consecutively
+numbered references. Nine tag pages and internal history links exist. History JSON/CSV contain
+all six additions with no raw citation marks. The country API has six segments and six populated
+context blocks; its final timeline score matches the exported exact score **23.5**. All six
+exported segment scores match the baseline, and the medical classification is Limited.
+Home, comparison and score exports contain no raw citation marks. `git diff --check` passed.
+
+Deleted `tmp/history-sint-maarten-gap-2026-10-23/site/` after verification. The earlier build
+from this batch was also deleted before rebuilding the final wording. Research files remain.