When an Electronic Formality Becomes a Barrier to Court
A documented Latvian e-justice case raises a question that every digital court system should be able to answer: when a filing has in fact been electronically signed, what justifies transforming it into the legal status of “not filed”?
Diks Ancveirs | Independent Legal Researcher
Access to Justice | Rule of Law | Effective Judicial Protection
6 October 2026
This article is based on primary procedural materials and the author’s SSRN and Zenodo academic editions. It does not allege personal misconduct by a judge and does not claim that a constitutional, Convention or EU-law violation has already been judicially established. Its purpose is narrower and more important: to examine whether the architecture of digital justice makes the path from technical validation to legal consequence sufficiently clear, reviewable and proportionate.
A question that cannot be dismissed as a “technical error”
On 6 October 2026, the Zemgale District Court addressed the admissibility of a civil claim. The decision records a crucial fact: the claimant had electronically signed an EDOC package containing the statement of claim and its attachments. The court nevertheless concluded that the statement of claim, as an autonomous procedural document, had not itself been separately signed with a secure electronic signature. The procedural consequence was severe: the claim was regarded as not filed.
This is where the case becomes larger than one dispute. Can the legal category “unsigned” be applied without further distinction to a situation in which an electronic signing act demonstrably occurred, the signer is identifiable, and the court itself acknowledges the existence of a signed EDOC package, but the dispute concerns the legal scope or granularity of that signature inside the container?
Courts are plainly entitled to require procedural documents to be signed. That is not the central challenge. A statement of claim institutes proceedings, and legal certainty about the act a person has adopted is a legitimate objective. Precisely for that reason, however, the legal classification must be exact.
The decisive question in digital justice is not only whether a signature exists. It is what was signed, how it was validated, how the law classified it, and why that classification justified the consequence imposed.
Four questions that should never be collapsed into one
Digital-signature disputes require at least four separate questions. First: does an electronic signature exist, and is it attributable to the relevant person? Second: what data, files, manifest or object references are actually covered by the cryptographic signature? Third: what legal effect does that signature receive under electronic-document law and eIDAS? Fourth: even if the signature is technically valid, does domestic procedural law require an additional document-level act of signature, and what consequence is proportionate if that requirement is not satisfied?
The distinction matters because “no signature exists” is not factually identical to “a signature exists, but the legal system considers its structural scope insufficient for this particular procedural document.” Those situations may ultimately receive the same procedural consequence under national law, but they should not reach that consequence through an opaque or imprecise classification.
From a technical fact to legal non-existence
The case can be understood as a three-stage transformation. At the first stage there is a factual event: electronic signing occurred. At the second stage comes legal classification: the package-level signature is treated as insufficient to sign the statement of claim as an autonomous procedural act. At the third stage comes the procedural consequence: the claim is regarded as not submitted.
That chain may be legally defensible. But its legitimacy cannot be established merely by repeating the final label. Between “the package was signed” and “the claim was unsigned” there must be a reconstructable technical and normative bridge.
Latvian civil procedure requires a statement of claim to be signed and attaches categorical consequences once a claim is classified as unsigned. This makes the preceding classification decisive. The deeper question is therefore not simply whether unsigned claims may be rejected. It is how an electronically authenticated filing is classified as unsigned in the first place.
eIDAS is not a magic key — but it cannot be made meaningless
Regulation (EU) No 910/2014, known as eIDAS, gives qualified electronic signatures the equivalent legal effect of handwritten signatures. EU law does not thereby abolish national procedural requirements or dictate every aspect of civil-court filing architecture.
The stronger legal argument is therefore not that eIDAS automatically overrides the Latvian Civil Procedure Law. The stronger question is whether the actual electronic signing event and its legal effect were sufficiently identified before the system imposed the harshest classification available: that the claim had never been filed.
A mature digital justice system must be able to explain the relationship between technical validity and procedural sufficiency. A technically valid signature may still be procedurally insufficient, but that conclusion requires a reasoned bridge, not a semantic shortcut.
Article 92 of the Satversme: form matters, but form is not an end in itself
Article 92 of the Latvian Constitution protects the right to defend one’s rights and lawful interests before a fair court. It does not create a right to disregard procedural rules. Courts need rules on jurisdiction, deadlines, signatures, fees and document form.
The constitutional difficulty arises in the relationship between the nature of a defect and the severity of its consequence. If identity, authorship, submission and integrity are objectively verifiable, the system should be able to explain what additional legitimate interest is protected by a separate file-level signature and why a less restrictive procedural response would not achieve equivalent certainty.
This is the point at which proportionality becomes central: legitimate aim, suitability, necessity, and a fair balance between orderly procedure and a person’s realistic ability to obtain a determination on the merits.
Article 6 ECHR and the boundary of excessive formalism
The European Court of Human Rights does not treat procedural formalities as inherently objectionable. They may protect legal certainty and the proper administration of justice. At the same time, access to court must be practical and effective, and excessive formalism can, in appropriate circumstances, become incompatible with Article 6.
The Grand Chamber’s approach in Zubac v. Croatia cuts both ways. It recognises procedural discipline and the responsibility of litigants to comply with rules, while also asking whether the restriction imposed remains proportionate. A fair analysis of the Latvian EDOC case must therefore acknowledge an adverse fact: the 6 October decision states that two earlier decisions had already explained the requirement for the statement of claim to be signed separately. That strengthens the argument that the requirement was foreseeable to this particular litigant.
But prior notice to one litigant does not answer the system-design question. Should a digital court system itself make the legally required signature structure visible before submission, especially where the consequence of getting that structure wrong is that the filing is treated as legally non-existent?
The deeper problem is the architecture of e-justice
In paper procedure, the legally operative document is usually intuitive: a sheet or set of sheets bears a handwritten signature. Containerised digital procedure separates the visible file, the EDOC container, the signature object, the manifest, the certificate, the timestamp and the transmission event. Law must decide which technical unit corresponds to the legal “document.”
A file-centric model can require every autonomous procedural file to carry its own signature. A container-centric model can treat a qualified signature over an identified container or manifest as authenticating its components. A hybrid model can distinguish constitutive procedural acts from evidentiary attachments. Any of these models may be defensible. What is not defensible in a mature digital system is leaving the user to infer the legally operative model from invisible technical architecture.
If the interface communicates “sign package” while procedural law later treats a central document inside that package as unsigned, the system contains a human-computer-law interface risk. The more sophisticated the technology becomes, the stronger the case for equally sophisticated validation, explanation and correction mechanisms.
If the system can detect the defect afterwards, why not before?
A powerful operational question follows from the case. If a court system can identify after submission that the principal pleading lacks the required individual signature, why should the same configuration not be detected before transmission?
A pre-submission check could state in plain language: “The statement of claim must carry its own qualified electronic signature. Signing only the EDOC package will not satisfy the required document-level signature.” Such a warning would not weaken procedural discipline. It would make compliance more reliable.
Get Diks Ancveirs’s stories in your inbox
Join Medium for free to get updates from this writer.
Digitalisation should not turn visible paper formalities into invisible metadata traps. A high-integrity system should prevent avoidable defects where possible rather than waiting for them to generate litigation about the filing process itself.
A system capable of detecting a signature-structure defect after filing should be asked why it could not warn the user before filing.
What Latvian e-justice should change
First, there should be one authoritative and technically precise public rule explaining when a package-level signature is sufficient and when an individual procedural document must be separately signed, with concrete examples for claims, applications, evidence and annexes.
Second, court portals and approved signing workflows should perform pre-submission validation for mandatory signatures. Third, the system should distinguish different defects rather than collapsing them into the single label “unsigned”: no signature, invalid signature, broken integrity, wrong signer, insufficient cryptographic scope, or procedural insufficiency are not the same technical condition.
Fourth, the court file should preserve a human-readable validation report together with machine-verifiable metadata showing the signer, certificate, timestamp, integrity result and the objects covered by the signature. Fifth, where the law permits and authenticity and intent are clear, a curable structural defect should trigger consideration of a correction mechanism before the filing is deprived of legal effect.
Sixth, automation must not obscure responsibility. The final legal classification must remain attributable to a human decision-maker and supported by a record of the technical facts and legal rule relied upon. Seventh, recurring e-justice questions should be capable of producing authoritative and publicly accessible guidance rather than remaining fragmented across non-appealable first-instance procedural decisions.
Traceability is the missing constitutional infrastructure
For high-stakes digital procedural acts, the minimum audit chain should be reconstructable in ordinary language: source, input, signature event, validation, human review, legal classification, procedural action, consequence, correction and review.
The source should preserve the original claim and attachments with stable integrity identifiers. The submission record should preserve the channel, timestamp and container identifier. The signature record should show the signer, certificate, signature type, timestamp and signing application. Validation should show what was actually covered and whether integrity checks succeeded.
Human review should identify who assessed formal compliance and what materials were reviewed. Legal classification should identify the exact rule and the precise defect. The procedural record should then show the action taken, its effect on filing date, limitation periods, interim protection or fees, and whether correction preserved or reset the legal position.
This is not an argument for replacing judges with software. It is the opposite. It is an argument for ensuring that digital procedure leaves a record from which human judicial judgment can be audited.
Why this concerns the whole of Latvia
Today the issue is EDOC. Tomorrow it may involve structured online forms, cloud signatures, API submissions, automatically generated bundles or AI-assisted court platforms. Technology is changing faster than traditional assumptions about what counts as a “document” and where the legally decisive signature resides.
This is therefore not a story about one file. It is about the State’s responsibility to build digital justice in a way that allows a person to foresee the legal consequences of an electronic act and allows a reviewer, months or years later, to reconstruct how a technical fact became a decision with consequences for a human being.
Public trust in a digital judiciary cannot rest only on the authority of the final decision. It also depends on a verifiable process: source, technical result, human review, legal classification, decision, consequence and correction.
Serious criticism must acknowledge the counterarguments
This case does not prove that the Latvian judiciary as a whole violates fundamental rights, and it does not justify allegations of personal misconduct. Serious rule-of-law analysis becomes stronger, not weaker, when adverse facts are stated openly.
A court has a legitimate interest in ensuring that a statement of claim, as the act instituting proceedings, is unambiguously adopted by the claimant. The 6 October decision refers to earlier explanations of the separate-signature requirement. Re-filing may also materially reduce the intensity of any fundamental-rights interference if it fully restores the litigant’s procedural position without loss.
Those counterarguments do not eliminate the systemic question. They sharpen it. If the State requires a highly specific digital signature architecture, that requirement should be clear, technically intelligible, capable of pre-submission verification and proportionate to the consequences attached to non-compliance.
Effective judicial protection must be effective in practice, not merely available in theory.
Conclusion: a digital State requires digitally mature procedural justice
Latvia has chosen digital public administration and digital court procedure. That is progress only if technology reduces uncertainty rather than creating new, invisible barriers.
The EDOC case asks a simple but difficult question: where a court recognises that electronic signing occurred but nevertheless classifies the procedural document as unsigned, can the system fully reconstruct and explain the path from the technical fact to the legal consequence?
If the answer is yes, the system should be able to show it. If the answer is no, the problem is no longer merely one litigant’s mistake. It is a question about the quality of e-justice itself.
The next standard for digital justice is not simply more electronic services. It is a system in which technical validation, legal classification, human accountability and correction form one transparent chain — before irreversible consequences occur.
Research and primary legal sources
SSRN Academic Edition: Electronic Signatures, Procedural Formalism and Effective Access to Court: A Latvian e-Justice Case Study
Zenodo Academic Edition: DOI 10.5281/zenodo.23194242
EU law: Regulation (EU) No 910/2014 (eIDAS)
Latvian law: Civil Procedure Law (Civilprocesa likums)
European Court of Human Rights: Zubac v. Croatia [GC], no. 40160/12, 5 April 2018.
Primary case anchor: Zemgale District Court, decision of 6 October 2026, Case №3–10/2049/43, ECLI:LV:ZRT:2026:1006.34069.1.L.
Methodological note: the underlying academic study distinguishes documented facts, legal characterisation, normative assessment and unresolved technical questions. A technical validation report for the original EDOC remains important before making any categorical claim about the exact cryptographic scope of the signature.
Suggested Medium tags
Access to Justice · Rule of Law · Digital Justice · Legal Technology · Human Rights
