Ask whether an electronic signature holds up in Guinea and you get three kinds of answer: a vendor’s yes, a sceptic’s no, and a lawyer’s it depends. All three skip the useful part, which is that the rule is short, public, and says something quite specific.
It fits in two articles of the Code civil. Here they are, and here is what they do.
Where the rule lives
The Code civil in force is Loi L/2019/035/AN of 4 July 2019, promulgated by Décret D/2019/222/PRG/SGG of 26 July 2019. Articles 1002 and 1003 are new in it: the code it replaced, from 1983, said nothing about electronic writing, because in 1983 there was nothing to say.
Where they sit matters as much as what they say. They are in the law of evidence. They are not a licensing regime and they do not appoint anyone to approve a product. They answer one question: when an electronic document is put in front of a judge, what is it worth?
Article 1002: electronic writing counts the same
Article 1002 : L’écrit électronique a la même force probante que l’écrit sur support papier, sous réserve que puisse être dûment identifiée la personne dont il émane et qu’il soit établi et conservé dans des conditions de nature à en garantir l’intégrité.
In English: electronic writing has the same evidential weight as writing on paper, provided the person it comes from can be duly identified, and it is created and kept in conditions that guarantee its integrity.
“La même force probante” — the same evidential weight. Not a lesser weight, not a supporting one. No discount for being electronic.
“Sous réserve que” — subject to. The equality is conditional, and there are exactly two conditions. The article names no technology, no format, no vendor, no certificate. Anything that delivers identification and integrity is inside the rule; anything that does not is outside it, however expensive it was.
Article 1003: what a signature is for
Article 1003 : La signature nécessaire à la perfection d’un acte juridique identifie son auteur. Elle manifeste son consentement aux obligations qui découlent de cet acte. Quand elle est apposée par un officier public, elle confère l’authenticité à l’acte. Lorsqu’elle est électronique, elle consiste en l’usage d’un procédé fiable d’identification garantissant son lien avec l’acte auquel elle s’attache. La fiabilité de ce procédé est présumée, jusqu’à preuve contraire, lorsque la signature électronique est créée, l’identité du signataire assurée et l’intégrité de l’acte garantie, dans des conditions fixées par décret.
Sentence by sentence: a signature identifies its author; it expresses consent to the obligations flowing from the act; affixed by a public officer, it makes the act authentic; when electronic, it consists in the use of a reliable identification process guaranteeing its link to the act it attaches to; and the reliability of that process is presumed, until proof to the contrary, where the signature is created, the signer’s identity assured and the act’s integrity guaranteed, under conditions fixed by decree.
Notice the order. Identification comes first, consent second, and only then the electronic form. Electronic signature is not its own species here. It is the same signature produced by a different means, judged on whether that means reliably identifies the person and stays bound to the document.
What “présumée, jusqu’à preuve contraire” actually does
This is the phrase people quote, and the one most often misread. It is worth being slow about.
A presumption is a rule about who has to prove what. It does not decide the case. It decides who has to open the argument.
Without it. You produce the signed document. The other side says the process was not reliable. You now have to show that it was: the identity check you ran, the seal on the file, the record of what happened and when. That is a winnable argument, and the work of winning it is yours.
With it. You produce the signed document, and its reliability is taken as established. The other side has to show it was not. The same evidence still carries you if the fight goes on, but you are not the one who has to start.
That is the whole effect. It moves a burden and buys you the opening position, which is worth a great deal in a real dispute and is not the same thing as winning.
Two things it is not.
Not a shield. “Jusqu’à preuve contraire” means what it says. Bring proof to the contrary and the presumption falls. A signature that meets the conditions is not unchallengeable. It is believed first.
Not a validity rule. A document does not become valid because a presumption applies, or void because it does not. Article 1002 stands on its own two conditions and mentions no presumption at all.
The six words at the end
“dans des conditions fixées par décret” — under conditions fixed by decree.
The presumption is not self-executing. It is offered on terms, and the article does not write those terms — it points at a separate text that has to say what “created”, “assured” and “guaranteed” concretely require.
For years that text did not exist. Article 1003 has been in force since 2019, and its closing clause pointed at a decree nobody had issued. The presumption was on the page and out of reach.
That is a smaller problem than it sounds. A dormant presumption removes an advantage; it does not remove the recognition. Article 1002 never depended on a decree, and article 1003’s own description of an electronic signature — a reliable identification process, linked to the act — did not either. What the missing decree withheld was the shortcut, not the signature.
So the honest answer, for the whole stretch from 2019 onward, was this: your electronic signature is evidence, and the burden of showing the process was reliable sits with you. Which is a very good reason to keep a signing record capable of doing that job unaided.
There is a second statute, with the same shape
Loi L/2016/035/AN of 28 July 2016 on electronic transactions came first, and anyone reading around this subject will meet it. Its article 21 admits electronic writing as evidence where its author “soit identifiée ou puisse être identifiée” — is identified, or can be identified. Its article 33 defines an electronic signature as “l’usage d’un procédé fiable d’identification”, the phrase the Code civil would later adopt almost word for word.
Its article 34 then does exactly what article 1003 does. It grants a rebuttable reliability presumption, but only to a higher tier it calls a signature électronique sécurisée, and it defers the conditions of that tier to a presidential decree. On the public record we have been able to consult, that decree was never issued either. Two parallel routes to a presumption, each written to wait on its own implementing text.
Two footnotes to it. Above it sits the ECOWAS Supplementary Act of 16 February 2010 on electronic transactions, which binds Guinea as an ECOWAS member — not the UEMOA instruments that circulate in regional summaries, since Guinea is not in UEMOA. And its article 35 gates recognition of foreign certificates behind a treaty, so what a certificate is worth where it was issued is a separate question from what it is worth in a Guinean file.
What to do with all of it
Build the file that would win without the presumption. Then a presumption, whenever it applies, is a bonus rather than a dependency.
Identify the signer, and record how. The condition is identification, not the presence of a signature image. The answer to “how do you know it was them” has to be in the file.
Seal the document so a later change is detectable. That is what “des conditions de nature à en garantir l’intégrité” asks for in practice: integrity you can demonstrate, not integrity you assert.
Keep the record with the document. A trail that lives only in a supplier’s dashboard is a trail you can lose.
What happened in August 2026
On 19 August 2026 a decree implementing article 1003 was announced on national television, setting the conditions the article had been waiting for since 2019.
Everything currently known about it comes from press accounts. It has no published reference number, no released official text, and no entry in the Secretariat General’s decree registry. That is not a reason to dismiss it, and it is a reason not to quote it yet as though it were settled law. The next post goes through what press accounts say it contains, and what remains unknown.
Nothing above depends on it. Articles 1002 and 1003 said what they say before 19 August 2026, and they say it still.
General information, not legal advice. Whether a particular document meets these conditions is a question for Guinean counsel, and contact reaches a person if you want to talk it through.