THE LABOUR COURT OF SOUTH AFRICA, JOHANNESBURG
Case no: JR 610/23
In the matter between:
GLORY MLAMBO Applicant
and
DELL COMPUTER (PTY) LTD First Respondent
PAUL PHUNDU N.O Second Respondent
THE COMMISSION FOR CONCILIATION, MEDIATION
AND ARBITRATION Third Respondent
Heard: 30 July 2026
Delivered: 7 August 2026
This judgment was handed down electronically by consent of the parties’
representatives by circulation to them via email. The date for hand- down is deemed to
be 7 August 2026.
JUDGMENT
PRINSLOO, J
Introduction
[1] The Applicant seek s to review and set aside an arbitration award dated 13
March 2023. The Second Respondent (the arbitrator) found that her dismissal
(1) Reportable Yes/No
(2) Of interest to other Judges: Yes/No
(3) Revised
____________ ______________
Signature Date
2
was fair in all respects and dismissed her case. The First Respondent (Dell)
opposed the application.
Background facts
[2] The Applicant was employed by Dell as an IT Analyst since 2002. Following a
performance improvement plan and incapacity inquiry for poor performance,
the Applicant was dismissed. She subsequently referred an unfair dismissal
dispute to the Third Respondent (CCMA).
[3] The matter was arbitrated , and the issue the arbitrator had to decide was
whether the Applicant had been unfairly dismissed. The arbitrator concluded
that the Applicant’s dismissal was both substantively and procedurally fair.
[4] The Applicant filed an application to review and set aside the arbitration award
in terms of section 145 of the Labour Relations Act
1 (LRA), which is opposed
by Dell.
[5] Before I deal with the merits of the review application, I deem it prudent to
address a few specific topics.
Affidavits in a review application
[6] In Elegant Line Trading 257 CC v MEC for Transport, Eastern Cape 2, the
Court confirmed that:
‘In motion proceedings the affidavits constitute both the pleadings and the
evidence and the issues and averments in support of the parties’ cases should
appear clearly therefrom. It is trite that an applicant must make out its case in
the founding affidavit which must contain sufficient facts in itself upon which a
court may find in the applicant’s favour.’
[7] It is trite that an applicant’s case must be made out in the founding affidavit.
In Airports Company of South Africa (SOC) Ltd v Tswelokgotso Trading
Enterprise CC3 the court held that:
‘Consequently, the applicant must set out sufficient facts in the founding
affidavit to disclose a cause of action, that is, the founding affidavit must be
1 Act 66 of 1995, as amended.
2 [2022] ZAECBHC 45 at para 2.
3 [2022] ZAGPJHC 410 at para 9.
3
self-contained. The replying affidavit (and in this instance the supplementary
affidavit) cannot be used to augment the applicant's case.’
[8] As was stated in Business Partners Ltd v World Focus 754 CC 4, in application
proceedings the affidavits constitute not only the pleadings but also the
evidence, and an applicant must therefore make out his or her case in the
founding affidavit and must stand or fall by the allegations contained therein.
[9] Affidavits in review applications serve two primary purposes: one, to define the
issues between the parties, and two, to place the essential averments and
evidence before the court and alert other parties about those. Rule 37
(previously 7A) of the Labour Court Rules (the Rules) provides for the delivery
of four sets of affidavits in review applications, to wit a founding,
supplementary, answering and replying affidavit.
[10] In recognition of the fact that the record obtained by an applicant in a review
application may reveal that averments made in the founding affidavit were
erroneously made or omitted, Rule 37 (20) and (21) (previously 7A(8)(a))
permits the applicant to deliver a supplementary affidavit within 5 days after
the transcribed record has been filed. This affords the applicant the
opportunity to supplement and amend the factual and legal grounds upon
which he or she relies considering the record. A weak founding affidavit can be
augmented by a supplementary affidavit.
5
[11] As a general principle, the applicant in a review application must make out his
or her case in the founding affidavit, as may be supplemented by a
supplementary affidavit, if necessary, after the transcribed record became
available. The supplementary affidavit may do no more than supplement the
grounds for review recorded in the founding affidavit or abandon any of them.
[12] The principle nonetheless remains that a case must be made out in the
founding papers. Its rationale promotes legal certainty. This is evident from the
founding papers. Its rationale promotes legal certainty. This is evident from the
approach adopted by the Constitutional Court in South African Transport and
Allied Workers Union and another v Garvas and others6, where it held that:
4 2015 (5) SA 525 (KZD) at para 8.
5 A Myburg, C Bosch Reviews in the Labour Courts,1st ed (2016) at pp 438 – 441.
6 2013 (1) SA 83 (CC) para 114.
4
‘Holding parties to pleadings is not pedantry. It is an integral part of the
principle of legal certainty which is an element of the rule of law, one of the
values on which our Constitution is founded. Every party contemplating a
constitutional challenge should know the requirements it needs to satisfy and
every other party likely to be affected by the relief sought must know precisely
the case it is expected to meet.’
[13] An applicant's pleadings contain the legal basis of the claim under which an
applicant has chosen to invoke the court's competence. In other words, not
only the formal terminology of the notice of motion, but also the contents of the
supporting affidavits must be interpreted to establish what the legal basis of
the applicant's claim is.
[14] Rule 37 (3) (previously 7A(2)(c)) provides that the notice of motion must be
supported by an affidavit, setting out the grounds upon which the applicant
relies to have the decision or proceedings reviewed, corrected or set aside.
[15] As to the requirement of setting out the grounds for review , it requires the
applicant to set out, with sufficient precision and detail, the grounds for review
and the bases upon which such grounds are relied on.
[16] The role of the reviewing Court is limited to deciding issues that are raised in
the applicant’s founding (and supplementary) affidavit. This was confirmed by
the Constitutional Court in CUSA v Tao Ying Metal Industries and Others (Tao
Ying)
7 where it was held that:
‘…the role of the reviewing court is limited to deciding issues that are raised in
the review proceedings. It may not on its own raise issues which were not
raised by the party who seeks to review an arbitral award. There is much to be
said for the submission by the workers that it is not for the reviewing court to
tell a litigant what it should complain about. In particular, the LRA specifies the
grounds upon which arbitral awards may be reviewed. A party who seeks to
grounds upon which arbitral awards may be reviewed. A party who seeks to
review an arbitral award is bound by the grounds contained in the review
application. A litigant may not on appeal raise a new ground of review. To
permit a party to do so may very well undermine the objective of the LRA to
have labour disputes resolved as speedily as possible.’
7 (2008) 29 ILJ 2461 (CC) at para 67.
5
[17] A party who seeks to review an arbitration award is bound by the grounds for
review contained in the review application, subject to one qualification, namely
that the Court is obliged to deal with a point of law apparent from the papers8.
[18] In short: it is critical that the legal foundation of the review application be
canvassed in the founding or supplementary affidavit and that it be linked to
the applicant’s grounds for review , which must be set out in sufficient
particularity. Abstract statements or generic conclusions, lacking detail, do not
constitute grounds for review.
[19] Grounds for review cannot be formulated for the first time in heads of
argument. In Northam Platinum Ltd v Fganyago NO and others
9 it was held
that:
'In my view the law is very clear that a ground for review raised for the first
time in argument cannot be sustained. The basic principle is that a litigant is
required to set out all the material facts on which he or she relies in
challenging the reasonableness or otherwise of the commissioner's award in
his or her founding affidavit.’
A section 145 review
[20] In National Bargaining Council for Road Freight and Logistics Industry v
Deysel N.O and Others 10 (Deysel) the Labour Appeal Court (LAC) considered
the test to be applied in review applications, pursued in terms of section 145 of
the LRA. The LAC held:
‘While the intention of the drafters of section 145 may have been to limit the
scope of review (as evidenced by the limitation of the grounds for review in s
145 to those recognised by section 33 of the Arbitration Act, the subsequent
enactment of section 33 of the Constitution, which guarantees the right to
administrative action that is lawful, reasonable and procedurally fair, has had
the effect of considerably expanding the scope for review. In Sidumo and
another v Rustenburg Platinum Mines Ltd and others (Sidumo), the
Constitutional Court held that a CCMA arbitration constituted administrative
Constitutional Court held that a CCMA arbitration constituted administrative
action but was not subject to the Promotion of Administrative Justice Act
(PAJA), and that section 145 of the LRA was to be interpreted as
8 Ibid.
9 (2010) 31 ILJ 713 (LC) at para 27.
10 (2025) 46 ILJ 1679 (LAC); [2025] 8 BLLR 790 (LAC)at paras 30-31.
6
encompassing the constitutional standard of reasonable administrative action.
The test for reasonableness to be applied in an application for review under
section 145 is whether the decision reached by the arbitrator is one that a
reasonable decision-maker could not reach.
It should be recalled that the issue that served before the court in Sidumo was
the “moral or value judgment to established facts and circumstances” applied
by a commissioner when determining the fairness of the penalty of dismissal.
The “threshold of reasonableness ” established by the judgment recognises
that in relation to the penalty of dismissal, value choices may differ in relation
to the same factual matrix but nonetheless fall within a range of decisions to
which a reasonable decision- maker could come. The metaphor of an elastic
band has been usefully employed to illustrate the applicable threshold – the
function of the review court is to determine the point to which the elastic of
reasonableness can stretch without snapping.’
[21] In MacDonald’s Transport Upington (Pty) Ltd v Association of Mineworkers
and Construction Union and others ,11 referring to Democratic Nursing
Organisation of SA on behalf of Du Toit and another v Western Cape
Department of Health and others,12 the LAC said that:
‘Since the advent of the Constitution of the Republic of South Africa 1996 (the
Constitution), the concept of review is sourced in the justifications provided for
in the Constitution and, in particular, that courts are given the power to review
every error of law provided that it is material; that is that the error affects the
outcome…
To recap, Navsa AJ said in Sidumo at para 105 that the review powers in
terms of s 145 “must be read to ensure that administrative action by the
CCMA is lawful, reasonable and procedurally fair”. Given that the section must
be interpreted to be in compliance with the Constitution, it would appear that
be interpreted to be in compliance with the Constitution, it would appear that
the concept of error of law is relevant to the review of an arbitrator’s decision
within the context of the factual matrix as presented in the present
dispute; that is a material error of law committed by an arbitrator may, on its
own without having to apply the exact formulation set out in Sidumo, justify a
review and setting aside of the award depending on the facts as established in
the particular case.’ (own emphasis)
11 (2016) 37 ILJ 2593 (LAC) at para 23.; [2017] 2 BLLR 105 (LAC).
12 (2016) 37 ILJ 1819 (LAC) at paras 21-22.
7
[22] In Deysel13 the LAC distinguished between review applications w here the
applicant seeks to review an arbitration award on the basis of a material error
of law committed by an arbitrator and where the review is sought because the
arbitration proceedings resulted in an unreasonable award.
[23] The difference is explained as follows:
‘Reading down section 145 to incorporate a requirement of reasonableness is
wholly appropriate in a case such as Sidumo, concerned as it was with the
exercise of a value judgment by an arbitrator in relation to fairness as a
penalty for misconduct, a judgment that by definition admits a range of
responses. Matters such as the present, where the administrative action in
issue involves a question of law that can produce a single correct answer, are
best understood and assessed when section 145 of the LRA is read as
suffused by the constitutional standard of lawfulness. Put another way, just as
the constitutional standard of reasonableness was found in Sidumo to have
suffused section 145, the constitutional standard of lawfulness does likewise.’
[24] What this approach recognises is that the right to review established by
section 145, where the applicant seeks to review an arbitration award on the
basis of a material error of law committed by an arbitrator, is not limited to
circumstances where the alleged error resulted in an unreasonable award.
The LAC confirmed that a material error of law is a discrete, substantive
ground for review under s ection 145 of the LRA and that a reviewing court,
when faced with what is alleged to be an error in law , does not have to assess
whether the arbitrator’s decision was reasonable.
[25] The LAC confirmed that:
In short: although a material error of law may previously have been viewed as
no more than a side car on the motorcycle of reasonableness, the
constitutional right to administrative action that is lawful requires that the
grounds for review established by s 145 of the LRA be understood as
grounds for review established by s 145 of the LRA be understood as
admitting a material error of law as a discrete, legitimate ground for review.
14
13 Id fn 10 at para 40.
14 Id fn 10 at para 43.
8
[26] Where the test to be applied is the one of reasonableness, t he applicant must
show that ultimately, considering the evidence placed before him or her, the
arbitrator arrived at an unreasonable result.
The review application: analysis
[27] The Applicant must make out her case in the founding affidavit, as may be
supplemented by a supplementary affidavit, and she has to set out, with
sufficient precision and detail, the grounds for review and the bases upon
which such grounds are relied on.
[28] In the Applicant’s founding affidavit, the grounds for review are recorded as
that the arbitrator erred when he accepted evidence, that he made a fatal
error, and that he made legal and factual errors. She stated that ‘another
person in the position of the second respondent would have arrived at a
different decision than that the second respondent had arrived at.’
[29] In her supplementary affidavit , the Applicant supplemented her grounds for
review and, in summary, her case is that the arbitrator ‘ made an error of law
and facts when he failed to take the body of evidence that was put before him ’
and ‘when he made the ruling, he made an error of law’. The remainder of the
averments contain statements such as that the management of Dell did not tell
her how to compile a report or did not show interest in her improvement, which
the arbitrator ignored to her prejudice. It is also contended that it was the
arbitrator’s ‘duty to ensure that expectations are communicated to the
employees in a fair manner’ and that he failed to do so in this regard.
[30] In Moraka v National Bargaining Council for the Chemical Industry and
others
15 the Court held that:
‘In setting out the grounds of review in his founding affidavit, the applicant did
not set out any factual basis for those grounds, but merely set them out in the
form of conclusions. Examples of this are the first two grounds of review he
mentions, namely:
“2.1 The Commissioner committed misconduct by making findings not
mentions, namely:
“2.1 The Commissioner committed misconduct by making findings not
justified on the evidence;
15 (2011) 32 ILJ 667 (LC) at paras 21 – 23.
9
2.2 Gravely misunderstood evidence presented before her;…” (sic)
The Labour Appeal Court has made it clear in the unreported case
of Comtech (Pty) Ltd v Commissioner Shaun Molony N.O. & Others (Case no
DA 12/05, dated 21 December 2007) that it is not sufficient for a party simply
to relate conclusions of law in the founding papers for a review application. A
party must set out the factual grounds on which it seeks to base its
review. While it may be excusable in a founding affidavit to state limited
grounds of review and in less detail, by the time an applicant has the record of
proceedings it must then make up for the deficiencies in the founding affidavit
and set out the factual basis for its grounds of review in full. When it came to
his supplementary affidavit, the applicant did not supplement or amend the
grounds of review set out in the founding affidavit, nor did he lay a factual
foundation for the grounds set out in the founding affidavit. On the approach of
the LAC in the Comtech case, no factual basis was provided for the review
application. It was only in his heads of argument that the applicant for the first
time set out a factual basis for his claim.
I am bound to follow the approach of the LAC in regard to the assessment of
the prospects of success and conclude that the applicant failed to provide any
factual basis for his grounds of review in his founding papers. Accordingly, it
is not necessary, on the basis of the Comtech approach, to consider the
merits of the case set out later, and for the first time, in the applicant’s heads
of argument.’
[31] In Tao Ying16 it was confirmed that the role of the reviewing court is limited to
deciding issues that are raised in the review proceedings. This Court, sitting as
a review Court, is not to embark on a fact -finding mission to find facts or
evidence to support the Applicant’s grounds for review. It is the Applicant’s
duty to tell this Court what evidence was not properly considered, why the
duty to tell this Court what evidence was not properly considered, why the
arbitrator’s findings were made in error and why the award was reviewable.
[32] In its opposing affidavit, the Respondent took issue with the Applicant’s
grounds for review and submitted that it was difficult to glean from the
founding and supplementary affidavits what the basis of the application is. The
Respondent submitted that the purported grounds for review are no more than
a list of evidentiary material which the Applicant alleges was not taken into
16 Id fn 7.
10
account, or they are an impermissible attempt to re- plead her version of
events, and further it amounts to an attempt to appeal the arbitration award.
[33] There are obvious difficulties with the Applicant’s grounds for review.
[34] First, and l et it be clear: this is an application for review and not an appeal.
This is a difference which this Court must maintain. In Bestel v Astral
Operations Ltd and others,17 the LAC considered the limited scope possessed
by this Court to review an arbitration award and accepted that an arbitrator’s
finding will be unreasonable if the finding is unsupported by any evidence, if it
is based on speculation by the arbitrator, if it is disconnected from the
evidence, if it is supported by evidence that is insufficiently reasonable to
justify the decision or if it was made in ignorance of evidence that was not
contradicted. The LAC held that:
‘…the ultimate principle upon which a review is based is justification for the
decision as opposed to it being considered to be correct by the reviewing
court; that is whatever this Court might consider to be a better decision is
irrelevant to review proceedings as opposed to an appeal. Thus, great care
must be taken to ensure that this distinction, however difficult it is to always
maintain, is respected.’
[35] Second, it is evident that the Applicant does not have a proper understanding
of what would constitute grounds for review. The grounds for review in her
founding and supplementary affidavits are not properly articulated. In fact, they
amount to no more than statements regarding evidentiary material she alleges
the arbitrator did not consider.
[36] The statements, or purported grounds for review, are extremely vague and
unsubstantiated. The Applicant did no more than launch a vague and
unsupported attack on the arbitrator’s findings and make bald allegations
which were not linked to the evidence or substantiated at all. The grounds for
which were not linked to the evidence or substantiated at all. The grounds for
review are rather legal conclusions or statements or simply the Applicant’s
personal disagreement with the arbitrator’s findings.
[37] The heads of argument filed by the Applicant were of little assistance, as they
contained no more than a paraphrasing of the averments set out in the
17 [2011] 2 BLLR 129 (LAC) at para 18.
11
Applicant’s affidavits. No submissions were made either on the
reasonableness test or the errors of law allegedly committed by the arbitrator.
[38] Third, as alluded to, the Applicant must make out her case in the founding and
supplementary affidavits. If the Applicant were to pursue the review application
on the premise that the outcome of the arbitration was unreasonable, the
review test as laid down in Sidumo and Another v Rustenburg Platinum Mines
Ltd and Others (Sidumo)
18 must be applied, and the Applicant’s grounds for
review in accordance with the Sidumo test should have been set out in her
affidavits.
[39] The Applicant must show that the arbitrator arrived at an unreasonable result.
In fact, the Applicant must show that the decision is one that a reasonable
decision maker could not make, based on the evidence presented. The test is
not whether the arbitrator came to the best decision or a decision acceptable
to the Applicant. A review is not an appeal. Unless the Applicant can show that
an error or misdirection caused the result of the award to be unreasonable, it
will be of no consequence, even if it exists. The ultimate question is whether,
holistically viewed, the decision taken by the arbitrator was reasonable based
on the evidence placed before him. This Court must consider the totality of the
evidence placed before the arbitrator and cannot embark on a process
whereby every minute detail of the arbitration award is assessed and analysed
- this will constitute a piecemeal approach which must be avoided.
[40] Where the proper right of review is one based on reasonable ness, that is the
case that must necessarily be pleaded. An award will only be set aside on
review if both the findings and the result are unreasonable. If the result is
capable of justification and therefore reasonable, it will not be interfered with
on review.
[41] In casu, t he Applicant did not make a single averment to the effect that the
on review.
[41] In casu, t he Applicant did not make a single averment to the effect that the
arbitrator’s findings were unreasonable, and she failed to show how the
arbitrator’s factual findings distorted the outcome and how it ultimately
rendered the arbitration award unreasonable. What the Applicant failed to
state was how the arbitrator’s factual findings (or errors) influenced the
outcome and how it rendered the arbitration award ultimately unreasonable.
18 (2007) 28 ILJ 2405 (CC) at paras 78 and 79.
12
[42] Fourth, it is evident from the affidavits that the Applicant’s case is premised on
errors that the arbitrator committed. Factual errors would fall under a
reasonableness review, which I already dealt with supra. What remains is the
Applicant’s grounds for review, relating to errors of law.
[43] Errors of law arise in relation to questions of law and refer to a wrong or
mistaken interpretation of a legislative provision. Q uestions of law are issues
determined by authoritative legal principles and include questions a court is
bound to answer in accordance with a particular rule or law, and questions as
to what the law is.19
[44] The LAC explained that a question of law involves a question that can produce
a single correct answer and is best understood and assessed when section
145 of the LRA is read as suffused by the constitutional standard of
lawfulness.
[45] The Applicant did not identify any specific error of law made by the arbitrator
and none was alluded to during argument.
[46] The Applicant simply failed to make out a case for review.
Costs
[47] This Court has a wide discretion in making a cost order, considering the
requirements of law and fairness.
[48] In addressing the issue of costs, Mr Kent for Dell argued that this is a hopeless
case and that Dell should not be expected to bear the costs of an application
that had no merit. Mr Sadike for the Applicant submitted that there should be
no costs order.
[49] Although this is a case where a cost order would be justified, I am mindful of
the fact that the Applicant is an un employed individual and that a cost order
would in all probability not be capable of execution. Therefore, I am of the view
that the interest of justice will be best served by making no order as to costs.
[50] In the premises, I make the following order:
19 Media Workers Association of SA and others v Press Corporation of SA Ltd (1992) 13 ILJ 1391 (A)
(Press Corp) 1396F-H, 1397J.
13
Order
1. The application for review is dismissed;
2. There is no order as to costs.
________________________
C. Prinsloo
Judge of the Labour Court of South Africa
14
Appearances:
For the Applicant : Advocate T M Sadike
Instructed by : Oosthuizen, Du Toit, Berg & Boon Attorneys
For the First Respondent : Mr J Kent of Solomon Holmes Inc Attorneys