On 25 September 2026, under heavy security at a Colombian National Police air base in Bogota, Geovany Andres Rojas — better known as “Spider” — was handed over to United States officials and flown out of the country. Rojas is accused of leading the Border Commands faction of the National Coordinator of the Bolivarian Army (CNEB), one of the dissident groups formed by former FARC fighters who rejected or abandoned the 2016 peace deal with Bogota. It is the first extradition of a senior commander of an illegal armed group under President Abelardo De La Espriella, who took office in August 2026 after campaigning against his predecessor’s negotiated-peace strategy. United States federal prosecutors accuse Rojas of drug trafficking and criminal conspiracy tied to alleged operations in Colombia and Ecuador dating back to 2017. The file has one explosive particularity: Rojas was arrested in February 2025 at a hotel in Bogota while, according to Reuters, he was participating in activities linked to the peace talks that the government of then-President Gustavo Petro was holding with illegal armed groups. A rebel leader seized at the negotiating table, then surrendered to a foreign power by the government that scrapped the talks — the case concentrates every hard question of modern extradition law: does an arrest connected to a peace process shield anyone, where does guerrilla warfare end and ordinary drug trafficking begin, and what would a court in Paris do with a file like this? This analysis answers from the standpoint of French extradition procedure, which offers one of the most protective judicial filters in Europe, and draws the practical lessons for anyone facing a cross-border surrender request defended from France.
I. Can a guerrilla leader seized during peace talks be extradited to the United States?
A. Why an arrest linked to peace talks does not by itself block surrender
The facts reported by Reuters on 25 September 2026 are worth setting out precisely, because extradition cases are won and lost on chronology. Rojas allegedly led the Border Commands, a faction of the CNEB operating along drug corridors. He was arrested in February 2025 in the Colombian capital while taking part in activities connected with the “Total Peace” policy of then-President Gustavo Petro, an attempt to negotiate simultaneously with several armed factions and with the ELN guerrilla after the 2016 settlement left dissident groups outside the peace architecture. That effort, Reuters recalls, was plagued by setbacks and violence. Abelardo De La Espriella, described as a 49-year-old right-wing lawyer, reversed course on taking office: he scrapped the talks with illegal armed groups, launched a military offensive to retake remote mountain and jungle territory, and on 25 September announced that his administration had already carried out 94 extraditions, declaring that Colombia would not be a refuge for drug traffickers wanted by international justice. The handover of Rojas, escorted by police officers before US agents at the Bogota air base, is the showcase operation of that doctrine.
For an international lawyer, the first reflex is to locate the applicable instrument. Between Colombia and the United States, surrender runs through classic diplomatic extradition under their bilateral treaty, not through the fast-track European arrest warrant created by the 2002 Framework Decision on the European arrest warrant and surrender procedures between Member States (Framework Decision 2002/584/JHA of 13 June 2002, CELEX 32002F0584, consulted on EUR-Lex). The distinction matters far beyond theory. Inside the European Union, surrender is a judicialised, deadline-driven mechanism with limited refusal grounds; outside it, extradition remains a sovereign act in which the requested state keeps a full political and judicial filter. Anyone comparing the two systems should therefore reason from the French domestic model, which governs every request France receives from a non-EU state such as the United States. Its opening rule is Article 696 of the French Code of Criminal Procedure, which provides, as verified on Legifrance: “En l’absence de convention internationale en stipulant autrement, les conditions, la procédure et les effets de l’extradition sont déterminés par les dispositions du présent chapitre. Ces dispositions s’appliquent également aux points qui n’auraient pas été réglementés par les conventions internationales.” (Article 696 of the Code of Criminal Procedure). In other words, the treaty applies first, and French statute fills every gap — a structure that gives the defence two successive layers of argument.
The second threshold question is whether the alleged conduct qualifies at all. French law applies double criminality and a seriousness floor. Article 696-3 of the Code of Criminal Procedure covers, first, all acts punishable as felonies under the requesting state’s law, and second, lesser offences “quand le maximum de la peine d’emprisonnement encourue, aux termes de cette loi, est égal ou supérieur à deux ans, ou, s’il s’agit d’un condamné, quand la peine prononcée par la juridiction de l’Etat requérant est égale ou supérieure à deux mois d’emprisonnement”, adding the absolute lock: “En aucun cas l’extradition n’est accordée par le gouvernement français si le fait n’est pas puni par la loi française d’une peine criminelle ou correctionnelle.” (Article 696-3 of the Code of Criminal Procedure). Large-scale cocaine trafficking and criminal conspiracy, as charged against Rojas for operations allegedly running since 2017, would clear that bar without difficulty in both systems. The seriousness of the charges, paradoxically, is what makes the file extraditable in the first place — and what pushes the defence toward the qualitative bars rather than the quantitative ones.
What, then, of the circumstances of the February 2025 arrest — seized while engaged in peace-process activities? Under French law, the conditions of arrest in the requesting state do not appear among the statutory refusal grounds, and the immediate French procedural answer to a surrender request is detention, not release. Article 696-11 of the Code of Criminal Procedure provides that once the extradition request is notified, if the prosecutor general decides not to leave the person at liberty, he presents him to the first president of the court of appeal, who “ordonne l’incarcération et le placement sous écrou extraditionnel de la personne réclamée à la maison d’arrêt du siège de la cour d’appel.” (Article 696-11 of the Code of Criminal Procedure). The French term matters: the person is held under an extradition detention order, a status distinct from ordinary pre-trial detention, and release under judicial supervision is possible only if representation at every step of the proceedings is sufficiently guaranteed. A lawyer defending a “Spider”-type client in Paris would therefore treat the peace-talks arrest as narrative and context — powerful before public opinion and potentially relevant to the political-offence analysis below — but would not present it as an autonomous legal shield. Promises or safe-conducts allegedly given during negotiations bind only to the extent they were formalised, and their breach is argued, in a French court, through the prism of the political-purpose exception and the fairness guarantees, not as a standalone immunity.
B. The political-offence test as a Paris court would apply it
The decisive qualitative filter is the political-offence exception. Article 696-4 of the Code of Criminal Procedure lists eight cases in which extradition “n’est pas accordée”, and its second ground provides: “Lorsque le crime ou le délit à un caractère politique ou lorsqu’il résulte des circonstances que l’extradition est demandée dans un but politique” (Article 696-4 of the Code of Criminal Procedure). The same article shelters French nationals judged by the nationality held at the time of the offence, offences committed on French territory, cases already finally judged in France, time-barred prosecutions, penalties contrary to French public order, trials without fundamental defence guarantees, and purely military offences. For a rebel commander, the entire defence converges on the second ground, under both of its branches: the political character of the acts, and the political purpose of the request.
Applied to the Rojas file, the test cuts both ways, and honesty requires saying so. On one side, the Border Commands are an armed faction born of a collapsed peace process; the arrest took place in the orbit of official negotiations; and the new administration openly presents the extradition as an instrument of its security doctrine, with 94 surrenders brandished as a political scorecard. A defence lawyer would argue that these circumstances colour the request with a political purpose within the meaning of the second branch. On the other side, the US prosecution targets drug trafficking and criminal conspiracy for profit since 2017 — archetypal ordinary criminality — and French courts have consistently refused to let an ideological backdrop launder profit-driven organised crime into political offences. The drugs-for-revenue structure documented by the Colombian authorities, with cocaine trafficking, illegal mining and extortion financing the armed groups, points toward common-law criminality even if the proceeds buy weapons. The lesson for any comparable file defended from Paris is procedural as much as substantive: the political argument must be evidenced act by act, document by document, because the chambre de l’instruction gives a reasoned opinion and can order further investigation before ruling. Article 696-15 of the Code of Criminal Procedure provides that where the requested person tells the prosecutor general he does not consent to extradition, “la chambre de l’instruction est saisie, sans délai, de la procédure”, that he appears before it within ten working days, and that “la chambre de l’instruction donne son avis motivé sur la demande d’extradition”, delivering that opinion within one month of the appearance unless further inquiries are ordered, with an unfavourable opinion required where “la cour estime que les conditions légales ne sont pas remplies ou qu’il y a une erreur évidente.” (Article 696-15 of the Code of Criminal Procedure). The defence therefore gets a full adversarial hearing before professional judges, with the right to demand supplementary evidence — the exact forum in which a political-purpose claim must be built, not merely asserted.
Recent French case law shows how seriously, and how strictly, that forum is policed — including in files with an overt political and military background. On 9 August 2023, the Criminal Chamber of the Court of Cassation ruled in a case, appeal no. 23-83.192 (ECLI:FR:CCASS:2023:CR01033), arising from an extradition procedure conducted at the request of the Ukrainian government before the Chambéry investigating chamber — a file unfolding while Ukraine was at war. The requested person sought release from judicial supervision after the investigating chamber issued an unfavourable opinion on extradition, while the prosecutor general had appealed. The Court of Cassation rejected the appeal, holding that “il s’induit des dispositions de l’article 696-17 du même code que la mainlevée de la mesure de contrôle judiciaire dont fait l’objet la personne réclamée ne saurait être acquise dès le prononcé d’un arrêt de la chambre de l’instruction qui a émis un avis défavorable à son extradition”, because under Article 569 of the Code of Criminal Procedure an appeal in cassation suspends execution of the court of appeal’s ruling until the Court of Cassation itself decides. The message for practitioners is twofold: even in a war-related extradition file, the Court applies the procedural code to the letter, and every interim measure — detention, supervision, release — is litigated blow by blow. A “Spider” defended in France would live under that same microscope from day one.
II. Wanted by Washington and arrested in Paris: how France would handle a “Spider” file
A. From extradition custody to the Prime Minister’s decree: the French procedure step by step
Imagine the counterfactual that interests every international reader of this column: Rojas, or a commander in his position, is arrested in Paris on a US extradition request founded on identical drug-trafficking and conspiracy charges. What happens next is neither the American nor the Colombian script, but a French procedure that stacks a judicial phase, a governmental phase and a judicial review of that governmental act — three bites at the file for the defence.
The judicial phase opens, as seen, with presentation to the first president and, failing sufficient guarantees of appearance, placement under extradition custody (Article 696-11). If the person refuses surrender, the investigating chamber of the Paris Court of Appeal is seised without delay and the adversarial examination begins: verification of identity, of the treaty basis, of double criminality under Article 696-3, and of each refusal ground under Article 696-4, from nationality — French nationals are not extradited, a protection assessed at the time of the offence — through prescription, prior French judgment, territoriality and military offences. The chamber’s reasoned opinion is determinative in one direction only. Article 696-17 of the Code of Criminal Procedure states: “Si l’avis motivé de la chambre de l’instruction repousse la demande d’extradition et que cet avis est définitif, l’extradition ne peut être accordée. La personne réclamée, si elle n’est pas détenue pour une autre cause, est alors mise d’office en liberté.” (Article 696-17 of the Code of Criminal Procedure). A final unfavourable opinion kills the request outright and frees the person as of right. But “final” is doing heavy work, as the 9 August 2023 Ukrainian-request ruling demonstrates: while the prosecutor general’s appeal in cassation is pending, the opinion is not final and interim supervision survives. Defence strategy must therefore plan for continued restraint measures even after winning the opinion, and be ready to relitigate liberty at every rung.
If the opinion is favourable, the file passes to the executive — and the executive is itself under judicial control. Article 696-18 of the Code of Criminal Procedure provides: “Dans les cas autres que celui prévu à l’article 696-17 , l’extradition est autorisée par décret du Premier ministre pris sur le rapport du ministre de la justice.” (Article 696-18 of the Code of Criminal Procedure). The same article builds two guillotine clauses that defence lawyers calendar immediately: if, within one month of notification of the decree to the requesting state, the person has not been collected by that state’s agents, then, absent force majeure, he is released as of right and may never again be claimed on the same grounds; and any action for excess of power against the decree must, on pain of foreclosure, be brought within one month, with a prior informal appeal to the administration having no suspensive effect on that deadline. In practice, the decree is challenged before the Conseil d’Etat on all classic grounds — error of law on the treaty, misapplication of Article 696-4, disregard of the investigating chamber’s reservations, breach of the European Convention on Human Rights — and the one-month foreclosure deadline makes speed the first virtue of the defence. Compare the Colombian showcase: a presidential announcement and a tarmac handover within the executive’s own tempo. In Paris, the executive signs but the administrative judge watches, and a missed collection deadline frees the prisoner permanently for that case.
One structural contrast deserves emphasis because it surprises American observers. Within the European Union, the Framework Decision of 13 June 2002 replaced political extradition with judicial surrender on strict time limits and a curtailed list of refusal grounds, including for a state’s own nationals in defined circumstances. Outside the Union, France reverts to the full classic grammar: no surrender of French nationals, systematic double criminality, political-offence scrutiny, decree, and Conseil d’Etat review. A Colombian commander wanted by Washington and stopped at Roissy would therefore face a slower, more layered and more litigable procedure than a suspect transferred between two EU capitals — slower, but also more protective, with each layer generating appealable acts.
B. The three shields: capital punishment, irreducible life sentences and specialty
Even where the charges qualify and no political bar applies, French extradition law deploys three substantive shields that any United States-bound file must survive. They are the features of the Paris forum that most astonish foreign prosecutors — and the points on which a French defence lawyer concentrates fire.
The first shield is the death penalty. France abolished capital punishment, and its courts will not deliver a person to a jurisdiction where he faces execution without an individualised, credible guarantee. The leading recent illustration is the Court of Cassation ruling of 30 January 2024, appeal no. 23-83.549 (ECLI:FR:CCASS:2024:CR00067), in an extradition requested by Tunisia where the Aix-en-Provence investigating chamber had given a favourable opinion, subject to an express reservation that any death sentence, if pronounced, would not be carried out. The Court of Cassation quashed the ruling in its entirety — “CASSE et ANNULE, en toutes ses dispositions, l’arrêt susvisé de la chambre de l’instruction de la cour d’appel d’Aix-en-Provence, en date du 31 mai 2023” — on a ground every extradition practitioner should memorise: “En se référant à des considérations générales relatives à l’existence d’un moratoire sur l’application de la peine de mort en Tunisie depuis trente ans et aux engagements internationaux renouvelés de l’Etat requérant en ce sens, qu’elle qualifie elle-même de symboliques, alors que ces assurances ne permettaient pas de s’assurer que la peine de mort, si elle était prononcée, ne serait pas appliquée spécifiquement à la personne de M. [V], l’arrêt attaqué ne satisfait pas, en la forme, aux conditions essentielles de son existence légale.” General moratoria and symbolic international commitments are not enough; the requesting state must give assurances that the penalty will not be applied to this specific person, issued by a competent authority, or the favourable opinion collapses. The statutory anchor is Article 696-4, which bars extradition where the requested act “est puni par la législation de l’Etat requérant d’une peine ou d’une mesure de sûreté contraire à l’ordre public français ;”. For a drug-trafficking file of the “Spider” type, where the United States charges carry severe custodial penalties, the Paris defence would force the requesting authorities onto this terrain from the first hearing: which penalties are actually incurred, which are excluded by binding assurance, and who signs that assurance.
The second shield extends the same logic to sentences that, without being death, extinguish all hope of release. The European Court of Human Rights held Belgium in violation of Article 3 of the Convention for extraditing a man to the United States where he faced an irreducible life sentence, finding that such a sentence was not reducible for the purposes of Article 3 (Trabelsi v. Belgium, HUDOC reference 001-146372). The lineage runs back to the foundational extradition judgment of the European Court of Human Rights, Soering v. the United Kingdom, application no. 14038/88, decided 7 July 1989 (HUDOC reference 001-57619), which established that surrender exposing a person to death-row conditions can breach the prohibition of inhuman or degrading treatment. French courts read these authorities together with the seventh refusal ground of Article 696-4 — no surrender where the person “serait jugée dans l’Etat requérant par un tribunal n’assurant pas les garanties fondamentales de procédure et de protection des droits de la défense” — to demand a forward-looking, individualised risk assessment: the actual sentencing exposure of this defendant, the real reducibility of the sentence under the requesting state’s law, and the concrete detention conditions he would face. A rebel commander facing a potential life term in the American federal system would put the reducibility question at the centre of his Paris defence, with expert evidence on federal sentencing and release mechanisms filed before the investigating chamber, not after the decree.
The third shield operates after surrender: specialty. The state that obtains a person may try him only for the acts for which extradition was granted, unless the surrendering state consents to an extension. The Bogota handover illustrates why the principle matters: the political narrative around the “Spider” file — dismantling trafficking chains “from cultivation to leadership” — is broader than any single indictment, and the temptation to add charges after surrender is the classic specialty scenario. French law treats the principle as a hard judicial control, including inside the European instruments. On 28 May 2024, the Criminal Chamber of the Court of Cassation, appeal no. 24-81.539 (ECLI:FR:CCASS:2024:CR00827), quashed a Paris investigating chamber ruling in a case chaining a Spanish surrender to Belgium with a French prosecution, recalling that a person transferred between Member States under the European arrest warrant mechanism cannot be detained for offences other than those motivating the transfer before the competent authority’s consent is obtained — the Court noted the breach of Article 18 of Council Framework Decision 2008/909/JHA on the transfer of sentenced persons — and laying down the rule of method: “il appartient à la chambre de l’instruction de s’assurer du respect du principe de spécialité.” The Court went further than words: it declared the detention baseless — “CONSTATE que M. [B] est détenu sans titre, dans la présente procédure, depuis le 8 février 2024” — and “ORDONNE la mise en liberté de M. [B], s’il n’est détenu pour autre cause ;”. Specialty is therefore not a diplomatic courtesy to be invoked after the fact; in Paris it is a justiciable, liberty-restoring control exercised by the investigating chamber itself, both before surrender, by framing precisely which acts the opinion covers, and after surrender, by policing extensions. Any lawyer negotiating or contesting a cross-border file builds the specialty perimeter into the earliest submissions, because what the surrender covers determines everything that follows.
Conclusion
The Bogota tarmac scene of 25 September 2026 will be remembered as the moment Colombia’s new security doctrine produced its first showcase: a rebel commander pulled from the shadow of a failed peace process and delivered to American justice. Whether that delivery survives the scrutiny of the United States courts — and of history’s judgment on trying warriors as traffickers — belongs to the coming months. What the file already teaches is structural. Classic extradition, unlike intra-European surrender, keeps the requested state’s full filter: treaty first, then statute; qualifying offences, then qualitative bars; judges, then the executive, then the judge of the executive. France’s version of that filter is among the most demanding in the world: a reasoned opinion of the investigating chamber within a month, automatic release on a final refusal, a Prime Minister’s decree challengeable within a one-month foreclosure deadline, individualised death-penalty assurances policed to the person, irreducible life sentences tested against Article 3, and specialty enforced by release orders. For governments, the lesson is that showcase surrenders buy headlines but also create precedents other courts will dissect. For defendants, the lesson is simpler and more urgent: in a cross-border file, the battle is fought in the requested state, in its language, under its procedure, and from the first hour of custody. That is where a Paris defence is built — act by act, assurance by assurance, deadline by deadline.