Introduction: a quiet question before a loud conclusion
There are moments in a nation’s constitutional life when the temperature rises quickly. A proposal appears, then a counter‐proposal, and soon the public square hums with confident predictions and determined certainties.
In such moments, the temperature rises faster than the understanding. Constitutional debate can easily become a contest of destinations rather than a journey through first principles. People rush to the “therefore” before pausing at the “why”.
Yet constitutions do not reveal themselves well to haste. They prefer to be approached with a certain steadiness, a willingness to sit with the quiet questions before addressing the loud ones. And the quietest question, the one most often drowned out when politics becomes animated, is also the most fundamental: from where does the authority to amend derive, and on what terms was it given?
Zimbabwe now finds itself at such a juncture. Parliament has introduced Constitutional Amendment Bill No. 3 (the “Bill”), proposing to extend the presidential term from five years to seven. More significantly, the Bill purports to apply that extension to the incumbent, expressly seeking to bypass the prohibition contained in section 328(7).
The public debate has understandably focused on consequences. But the issue examined here is narrower and more exacting. It is not whether longer presidential terms are desirable. Reasonable people may disagree on the merits of tenure. Nor is it whether constitutional reform, as such, is legitimate; it plainly is. The question is whether the Constitution permits the present office-holder to benefit from such a change. That inquiry turns not on political preference, but on the bedrock of constitutional structure.
Parliament and the Nature of Delegated Power
Zimbabwe’s Parliament is not sovereign in the classical sense. It is not the ultimate source of law. It does not stand above the Constitution. Rather, its powers exist because the Constitution establishes them and exist only within the parameters that the Constitution defines.
Section 117(1) provides that legislative authority derives from the people and is vested in Parliament “in accordance with the Constitution.” That phrase is foundational. Parliament’s authority does not exist apart from the Constitution. It exists through it.
Section 117(2)(a) continues: “the Legislature has the power to amend this Constitution in accordance with section 328.”
Those words are deliberate and are doing more than prescribing a procedure. They define the very source and scope of the amending power. The Constitution does not say that Parliament may amend the Constitution, full stop. It says Parliament may amend in accordance with section 328. That qualification is not ornamental. It is constitutive.
The power to amend is therefore not freestanding. It is conditional. Compliance with section 328 is not an external requirement appended to an otherwise complete power. It is the very condition upon which the power to amend comes into existence at all.
Delegated authority always carries boundaries. Those boundaries do not merely regulate how power is exercised; they define whether it exists at all in a given case. This distinction is subtle but important. Where a power is conditional, non-compliance does not produce an irregular exercise of power. It produces an absence of authority.
With that framework in mind, attention must turn to section 328.
Of Term Limits and Term Lengths
Section 328
Section 328 is often described as a procedural roadmap, a checklist of notice periods, and thresholds for how amendments must be made. That description understates its significance. Section 328 is the constitutional architecture through which amendment power is channelled and confined. It prescribes process and also delineates limits. While some subsections regulate mechanics, subsection (7) addresses substance and consequence.
It provides that: “Notwithstanding any other provision of this Constitution, an amendment to a term-limit provision, the effect of which is to extend the length of time that a person may hold or occupy any public office, does not apply in relation to any person who held or occupied that office, or an equivalent office, at any time before the amendment.”
The language is deliberate, precise, and absolute. It does not prevent Parliament from altering presidential term lengths in general. Nor does it freeze the Constitution in its present form. What it does, however, is separate constitutional change from personal benefit.
To understand the breadth of this protection, two distinct elements demand attention:
First, the safeguard is not confined to the numerical cap on terms. By deploying the phrase “an amendment to a term-limit provision,” the Constitution directs its focus to the entirety of the provision governing tenure.
Secondly, the provision focuses expressly on effect. It captures any amendment “the effect of which is to extend the length of time” in office.
The Constitution is properly concerned not with the nomenclature of an amendment, but with its substantive effect on tenure.
Under the current framework, the maximum duration of the presidency is fixed at ten years. This duration is not a collateral detail; it is intrinsic to the limit itself. A “term limit” is not a bare numeral; the number ‘two’, existing in isolation. It is a composite concept: “two periods of five years.”
To extend each term to seven years while maintaining the two-term cap is to shift the maximum tenure from ten years to fourteen. One cannot increase the length of the units without enlarging the total of the sum.
Consequently, to increase the duration of each term is, as a matter of inescapable logic, to enlarge the total permissible tenure. The Constitution’s use of the word “effect” in section 328(7) ensures that such a practical enlargement cannot be disguised as a mere administrative adjustment.
It is therefore legally incoherent to suggest that extending the period of service leaves the term limit untouched. While the “count” may remain at two, the tenure is enlarged. It follows that an amendment which lengthens each term, while leaving the number of terms untouched, nevertheless extends the total time a person may hold office. That is precisely the outcome subsection (7) addresses. The deliberate inclusion of the language of “effect” prevents circumvention by drafting ingenuity. The safeguard is directed at substance.
Moreover, if subsection (7) could be bypassed by simply relabelling an extension of tenure as a modification of term length, the safeguard would be rendered illusory. The very mischief the provision seeks to prevent – incumbents leveraging their power to secure personal benefit – could be achieved through mere semantic adjustment. Constitutional protections of this gravity are not drafted for evasion; they are drafted to endure.
The logic of section 328(7) is both forward-looking and structural. It prevents constitutional reform from being used as a tool to recalibrate incumbency, ensuring that altered terms of office operate prospectively, not retroactively.
The rationale for such a rule is manifest. Incumbents are uniquely positioned to shape the constitutional weather: they control legislative timetables and frame the discourse of reform. Subsection (7) is therefore prophylactic, not punitive. Its purpose is to maintain the integrity of the constitutional compact by separating the rules governing the future from the interests of the present office-holder.
Section 91 v Section 95
It may be suggested, however superficially, that the true “term-limit provision” is confined to section 91(2), which disqualifies a person from election after two terms. On this view, section 91 fixes the limit, while section 95 merely describes the collateral mechanics of a term’s duration.
This distinction cannot withstand scrutiny. Section 91 speaks in the language of disqualification; section 95 in the language of tenure. Yet the two are inextricably linked. Section 91 caps the number of terms, while section 95 provides the substantive content to the word “term.”
A limit of “two terms” is not an abstract numeral; it is a reference to two constitutionally defined periods of time. Duration and number are not competing concepts; they are the twin variables by which maximum tenure is calculated. To alter either is to shift the constitutional boundary.
To isolate section 91 as the sole “term-limit provision” is to divorce the word “term” from the very provision that defines it. The Constitution does not treat these sections as strangers; it treats them as the constituent parts of a single, unified tenure scheme. To enlarge the duration of the unit is, by definition, to enlarge the limit of the whole.
The Memorandum
If the above were considered mere conjecture, the Explanatory Memorandum to the Bill provides dispositive clarity. It expressly justifies the transition from five-year to seven-year terms as a means to “to eliminate election mode toxicity and allow sufficient time for project implementation while promoting stability.” The amendment is thus expressly justified on the basis that five years is insufficient time in office and that seven years would provide more.
This is not a description of a minor administrative adjustment. It is an explicit admission of an intent to enlarge tenure. Section 328(7) is triggered whenever an amendment has the “effect” of extending the length of time an official may hold office. By the government’s own admission, that is precisely what is being done. The Constitution’s concern is with the substance of the outcome, and in this instance, the outcome is unmistakable.
Judicial Guidance: Mupungu
The structural integrity of this argument is confirmed by the Constitutional Court’s analysis in Mupungu v Minister of Justice, Legal and Parliamentary Affairs [2021] ZWCC 7. There, Patel JCC observed:
“By way of contrast, the Constitution abounds with a myriad of provisions that unquestionably constitute specific term limit provisions within the parameters of s 328. First and foremost, there is s 95(2) which expressly stipulates that the term of office of the President is 5 years and coterminous with the life of Parliament… Consequently, an amendment to any such provision, the effect of which is to extend the length of time that a person may hold or occupy the public office in question, falls squarely within the ambit of s 328(7).”
While these observations were obiter dicta, they were far from incidental. They represent a rigorous mapping of the constitutional terrain. In matters of high constitutional importance, such deliberate judicial analysis, particularly where it is closely tied to the constitutional text, is entitled to considerable persuasive weight. The passage is notable for two reasons:
First, it expressly identifies section 95(2) – the very provision governing presidential tenure – as a “specific term limit provision” for purposes of section 328.
Secondly, it affirms that an amendment extending the duration of time in office triggers the subsection (7) safeguard.
By characterizing the five-year duration as a “specific term limit provision,” the Court recognised that length and limit are functionally indistinguishable. The logic is mathematically precise: if a person is limited to two terms, and the duration of those terms is enlarged, the “length of time” that person occupies office is extended. Section 328(7) is triggered not by the name of the section being amended, but by the material effect on tenure. Mupungu serves as a judicial mirror, reflecting what the text already makes plain: the Constitution separates the power to reform the office for the future from the power to enlarge the mandate of the present.
Literal Meaning and Constitutional Interpretation
The interpretive task required here is disciplined, not creative. Zimbabwean courts have consistently held that constitutional provisions must be given their plain grammatical meaning unless doing so would produce an absurdity (see Mavedzenge v Minister of Justice, Legal and Parliamentary Affairs CCZ 2018-05).
There is no ambiguity in sections 117 and 328. Section 117 confers the power to amend only in accordance with section 328. Section 328(7) withholds retrospective benefit from incumbents where tenure is extended.
The words are plain. The structure is coherent. No special interpretative gymnastics are required; there is no need to mine Hansard or to invoke abstract constitutional theory. There is no need to consult foreign jurisprudence.
The answer lies not in implication, but in articulation. The Constitutional text itself resolves the issue.
The Explanation at the Time of Adoption
This literal approach is not merely a matter of legal preference; it is the basis upon which the Constitution was presented to the nation.
When the draft was put to the public in 2013, the Constitutional Parliamentary Committee (“COPAC”) issued a summary to ensure the electorate made an “informed decision.” That summary was explicit.
It stated that while the President would serve two five-year terms, “an amendment to a term limit to extend the term does not extend the term of any person in office before the amendment.”
This was not a casual simplification but a definition of the mischief the law was designed to preclude.
It served as a public guarantee that the total period of service was fixed and could not be stretched; whether by increasing the number of terms or lengthening their duration, by those the law was intended to restrain.
By using such plain language, COPAC confirmed that Section 328(7) was a substantive part of the constitutional structure. To interpret the section otherwise today is to suggest that the public were invited to endorse a text that said one thing, while the law secretly permitted another.
The mandate was given on those terms, and the Constitution does not allow them to be changed after the event.
The Difficulty with “Notwithstanding”
The Bill’s proposal that the extension of presidential tenure shall operate “notwithstanding” section 328(7) is more than a mere drafting choice; it is a fundamental misunderstanding of the nature of delegated power.
The drafters recognise section 328(7) as a hurdle and seek to legislate past it. However, a constitutional constraint does not evaporate simply because it has been described as inconvenient.
The fatal flaw in the Bill’s logic is a classic jurisdictional bootstrap. In the Zimbabwean constitutional order, Parliament is not a sovereign body possessing original, inherent powers.
It is a creature of the Constitution, and its legislative authority, specifically its power to amend, is derived and conditional. Section 117(2)(a) confers the power to amend only on the condition that it is exercised “in accordance with section 328”.
To invoke section 117 while simultaneously repudiating section 328(7) is to attempt to lift oneself up by one’s own laces.
It is an exercise in circularity:
• Parliament may only amend the Constitution by following the mechanism of section 328.
• Section 328(7) is an integral part of that mechanism, specifically withholding retrospective benefits from incumbents.
• If Parliament ignores the condition (the limit on incumbency), the power to amend is never engaged in the first place.
One cannot rely on the authority of section 117 while rejecting the very channel through which that authority flows. This is not a technicality. It goes to the heart of jurisdiction. A power conferred subject to a condition cannot be exercised by the simple expedient of ignoring that condition.
If the condition for the power’s existence is discarded, the result is not an “amendment” at all. It is a nullity. Parliament may redesign the office of the Presidency for the future, but it lacks the competence to grant itself a self-exemption from the rules of its own creation.
The Unanswered Question: On What Authority?
Once this structure is understood, a decisive question emerges. If Parliament is not amending in accordance with section 328, on what constitutional authority is it acting?
It cannot be section 117, because that section is tethered to section 328.
It cannot be residual sovereignty, because Parliament does not possess it.
It cannot be any other provision, for none exists that confers a freestanding power to amend.
The Bill therefore rests on no identifiable source of constitutional power. The use of the word “notwithstanding” is not incidental. It is a tacit admission that section 328(7) stands in the way. Indeed, the attempt to override subsection (7) only confirms its operative force.
What ultimately resolves the matter is not political theory, nor comparative jurisprudence, nor even judicial gloss. It is structure. The Constitution confers upon Parliament the power to amend, but only through the mechanism it prescribes. That mechanism contains within it a safeguard which insulates constitutional reform from personal advantage.
To say that Parliament may amend “in accordance with Section 328” is to say that it may amend subject to every constituent element of that section, including the prohibition on retrospective benefit. Sever that condition, and what remains is not a broader amending power. It is a legal vacancy. The attempt to exclude subsection (7) is therefore not an exercise of amendment within the constitutional framework. It is an attempt to alter the framework while standing outside it.
The Referendum and its Limitations
It may be argued that any constitutional difficulty disappears if an amendment is passed by a two-thirds majority in Parliament and approved in a national referendum. That argument rests on subsection (6), on the assumption that once the required majorities are obtained, the amendment is beyond objection.
However, that assumption confuses two separate questions. The first is how a Constitutional Bill is enacted; the second is what that Bill is permitted to do once it exists. Subsection (6) answers the first question. Subsection (7) answers the second.
Subsection (7) begins with the words: “Notwithstanding any other provision of this section.” Those words are not decorative. They establish a hierarchy. They mean that even if every other rule in the section is followed, including the rule for a referendum, the limit in subsection (7) remains controlling. Even where a referendum is held and won, an amendment that extends tenure “does not apply” to the person currently in office. Popular approval cannot grant a power that the Constitution has expressly withheld.
The point is straightforward. A referendum may validate the passage of an amendment, but it does not alter the limitation placed on its application.
Subsection (8) reinforces this structure. By preventing subsections (6) and (7) from being amended together or even presented together in the same referendum the Constitution ensures that the safeguard in subsection (7) cannot be neutralised by the procedural mechanism of a vote or referendum. The route to amendment is kept strictly separate from the protection against retrospective benefit.
Unless subsection (7) is itself validly amended, it continues to bind. The present proposal does not seek to amend subsection (7); it seeks to operate “notwithstanding” it. The Constitution provides no such path. A referendum may alter the tenure of the office for the future. It cannot enlarge an existing mandate.
Electoral Mandates and Retrospective Enlargement
Beyond the technicalities of section 328, there is a foundational democratic dimension to consider. Presidential authority is not an absolute grant; it is conferred personally, through election, on defined terms and for a specified duration. The President was elected in 2023 for a five-year term. That mandate was sought on that basis. It was granted on that basis.
That authority is complete, not provisional. It is bounded by the very terms upon which it was conferred. It cannot be extended after the fact, even by parliamentary or popular initiative, without a fresh electoral mandate conducted under the revised constitutional framework. Constitutional amendment may reshape the architecture for the future, but it cannot retrospectively enlarge a mandate already conferred.
To suggest otherwise is to unsettle both constitutional form and democratic legitimacy. Consent given for a five-year term cannot, by legislative fiat, be stretched to encompass seven. Only a fresh mandate, obtained in a new election, could confer legitimacy upon an extended tenure. In the absence of such a mandate, an extension is not a reform; it is a unilateral variation of the constitutional bargain.
Consent given for one term cannot silently expand into another.
A Restrained Comparative Insight: South Africa
Although the Constitution’s text alone is decisive, it is worth observing that South Africa’s constitutional jurisprudence reflects similar discipline. In Premier of KwaZulu-Natal v President of the Republic of South Africa and Others [1995] ZACC 10, the Constitutional Court articulated a principle that resonates deeply with the present crisis:
“There is a procedure which is prescribed for amendments to the Constitution and this procedure has to be followed. If that is properly done, the amendment is constitutionally unassailable. It may perhaps be that a purported amendment to the Constitution, following the formal procedures prescribed by the Constitution, but radically and fundamentally restructuring and re-organizing the fundamental premises of the Constitution, might not qualify as an “amendment” at all.”
The relevance lies not in the importation of foreign doctrine, but in the recognition of a universal constitutional truth: the authority to amend is itself a regulated power.
Section 328(7) in Zimbabwe serves precisely this function. It acts as a constitutive boundary, marking the line between a legitimate amendment and a fundamental alteration of the constitutional compact in favour of incumbency. When Parliament seeks to bypass this boundary, it is not “amending” the Constitution; it is attempting to re-organize its fundamental premises without the requisite authority. The South African experience confirms that courts must be the ultimate guardians of this distinction.
Conclusion: Drawing the Threads Together
Against this background, the legal reality is stark. The Constitution is not a menu of optional provisions from which inconvenient clauses may be omitted. The power to amend exists only within the framework that defines it.
The logic follows a chain from which there is no escape:
Parliament may amend the Constitution under section 117 only ‘in accordance with’ section 328.
Section 328(7) provides that tenure-extending amendments do not apply to an incumbent.
A Bill that purports to exclude section 328(7) is not made in accordance with section 328.
Consequently, such a Bill is not authorised by section 117.
A Bill without the authority of section 117 is a constitutional nullity.
This is not a matter of political friction; it is a matter of constitutional permission. To honour section 328(7) is not to resist reform; it is to respect the very terms upon which the power to reform was granted.
Amendment is possible. Reform is permitted. But the self-extension of power through the strategic displacement of constitutional safeguards is a jurisdictional impossibility. This conclusion is neither radical nor partisan. It is structural. The Constitution may be amended. It may not be bypassed.
The discipline required to reach this conclusion is not unique to constitutional law. It is the same discipline required whenever an exercise of power, whether regulatory, administrative, or commercial, seeks to outrun its legal anchor.
In our practice, we find that complex challenges are rarely resolved by assertion alone, but by the same structural analysis applied here: returning to the text, identifying the source of authority, and respecting the limits of the grant.
Whether at the level of the state or the institution, clarity is the product of that interpretive rigour. It is the standard we apply so that when our clients move forward, they do so on a foundation that is legally secure.
This article was written by Irvine Chiwara, a legal practitioner in Zimbabwe, and a principal at FLP.
