by invoking the presumption under Section 114A of the Evidence Act 1950 against the 1st Defendant.” [71] In my view, the two grounds relied on by the plaintiff failed to prove publication by the 1st defendant. My reasons are as follows. Plaintiff’s 1st ground - “that based on the facts and law, the irresistible conclusion and/or irresistible inference is that the 1st defendant jointly and/or severally published the Poison Pen Letter to third parties.” [72] The plaintiff has cited numerous cases where the courts had considered whether the human defendants in those cases had published anonymous letters which were defamatory of the plaintiffs in those cases. [73] Looking at the totality of the facts and evidence presented in those cases, the courts there had made inferences and concluded that the human defendants there had indeed published the anonymous letters which were defamatory of the plaintiffs in those cases. The cases cited by the plaintiff are Pan v Cheng; Zhou v Cheng [2021] NSWSC 30 (Supreme Court, New South Wales), Amanatidis & Anor. v Darmos [2011] VSC 163 (Supreme Court, Victoria), Richard Alan Parsons v. Elizabeth Garnett & 2 Ors. [2022] EWHC 3017 (KB) High Court of England & Wales, Maui Ashley Solomon v David James Prater [2021] NZHC 481 (High Court, New Zealand), and Stanislaus J Vincent Cross v Ganesan Vyramutoo & Anor. [2020] 10 CLJ 263 (High Court). [74] These cases cited by the counsel for the plaintiff sets out the law on how a court can determine that a particular human defendant is responsible for an anonymous defamatory letter. But it is a far cry from the facts of our case here. [75] In our case, those cases cited by the plaintiff would be relevant and indeed pertinent if it’s against the late 2nd defendant, who has since died and the suit abated. [76] But these cases are not applicable to the claim against the 1st defendant as the plaintiff had not identified and pleaded the name(s) of the employee(s) of the 1st defendant who had supposedly published the Poison Pen Letter for and on behalf of the 1st defendant. Neither has the plaintiff led any evidence to prove an employee(s) of the 1st defendant was authorised to publish the Poison Pen Letter for and on behalf of the 1st defendant. Is Metadata analysis reliable? [77] I will now examine the plaintiff’s claim that the metadata properties of the Microsoft Word copy of the Poison Pen Letter dated 28-09-2018 [78] On 30-10-2018 a Microsoft Word copy of the Poison Pen Letter dated 28-09-2018 was sent via email from one dk@mustapharaj.com to one michelle_sunita@hotmail.com. This Microsoft Word copy of the Poison [79] The plaintiff’s defamation suit against the 1st defendant is premised solely on the fact that the “Author” and “Last Modified By” metadata properties of the MACC.docx refer to a user name “Leaderonomics”, which is the company name of the 1st defendant. [80] Thus, the plaintiff alleges that the “Metadata Shows Poison Pen Letter originated from the 1st Defendant (‘Leaderonomics’).” [See Enclosure 112 plaintiff’s Written Submissions paragraph 75]. [81] There are two flaws in the plaintiff’s submissions. [82] Firstly, the 1st defendant is a corporate entity and not a human. A company, not being a natural person, cannot act on its own. It can act only through the agency of natural persons. [83] The Federal Court in Tengku Dato’ Ibrahim Petra bin Tengku Indra Petra v Petra Perdana Bhd and another appeal [2018] 2 MLJ 177, held: - “[103] A company is an artificial person and has no physical existence. Its legal existence is recognised only by reason of the Act. A company, not being a natural person, cannot act on its own; it can act only through the agency of natural persons. As stated by Cairns LJ in Ferguson v Wilson (1866) LR 2 Ch App 77 at p 89 ‘the company itself cannot act in its own person, for it has no person’.” [emphasis added] [84] In Christopher Grant v Teh Beng Leong & Anor [2019] MLJU 1574 (High Court) the court said: - “[10] The plaintiffs claim against the 1st defendant for negligence is clearly misconceived. The 1st defendant is a body corporate or an artificial legal person created by written law. As such it can only act through living persons and liability visited upon it vicariously for the acts of those persons such as its employees or agents. The 1st defendant is therefore incapable of committing the tort of negligence. On the other hand, any person who commits the tort of negligence is personally liable to the person he causes injury or damage to.” [Emphasis added] (Court of Appeal for Ontario)]. But this is not the plaintiff’s pleaded case. The plaintiff’s pleaded case is not that an authorised employee of the 1st defendant published the Microsoft Word copy of the Poison [86] The plaintiff’s pleaded case is that the 1st defendant itself published the Microsoft Word copy of the Poison Pen Letter dated 28-09-2018 Metadata properties analysis is unreliable [87] Secondly, the plaintiff’s claim that the metadata properties of the Poison Pen Letter is contradicted by the plaintiff’s own two forensic experts. Cross-Examination of PW3, Akash Rosen (Notes of Proceedings page 148-150) DHR And in all the copies that you analysed, Mr. Akash, does the 1st Defendant appear in any of those documents that you analyse? PW3 Yes. As the “author” and the “last saved”. DHR Only that right? DHR Nothing more? There is no link to show that it originated from a server or from a website? PW3 No. DHR Or some other IP address that can be traced to the 1st Defendant? PW3 No. There is no way you can get those IP addresses. … DHR There is also no evidence that it originated from an email address or an Internet account owned by the 1st Defendant. Correct? Defendant. PW3 From the document, no. [89] Under cross-examination, the plaintiff’s 2nd expert witness PW-4 Steven Wilkinson admitted the metadata properties should only be taken at face value, namely what is the user name at the time the MACC.docx was created and last modified. These metadata properties do not prove the identity of the actual user behind the user name “Leaderonomics” - Cross-Examination of PW4, Steven Wilkinson (Notes of Proceedings page 184-185, 194-195) DHR So, when you say you understand “author” to be writer, it’s basically a system feature that tells you to form that conclusion, correct? PW4 It’s a feature of the Microsoft application, yes. DHR It doesn’t necessarily mean that the name that appears is the actual writer? PW4 I couldn’t put a writer’s fingers on the keys, no. DHR Taking your statement at its highest, would you say that an organization wrote the document that you were forensically analyzing? PW4 I would say that that user name would indicate an organization. DHR That is the highest it goes right? That is the highest possible inference that you can draw from your statement? PW4 I can infer that, that Leaderonomics is, would indicate to me as an organizational name… … DHR So, the First Defendant’s expert is saying that the “author” and “last modified” metadata contained in the Word document has no value when identifying or verifying authorship or publication. And looking at your response, basically you agree that there is no, there’s little value in identifying or verifying authorship, or in this case writership, and publication? PW4 No, they actually commented the “author” and “last modified” metadata is, has no value. All metadata can be altered or changed in some way. And it is a generally held opinion within the forensic circles that metadata should not be relied on, on its own to prove any particular point, it merely gives an indication. [Emphasis added]. [90] I find PW4 Steven Wilkinson’s testimony honest, candid and useful. [91] The 1st defendant’s expert DW3 has also shown that the “Author” and “Last Modified By” metadata properties can be specified, edited, and substituted by any Microsoft Word user on any electronic device using the Microsoft Word application - Cross-Examination of DW-3, Tan Keh Win [Notes of Proceedings page 328-337] EK Okay, so how are they specific seeing as you have not studied that document? DW3 Because I saw the initial report by Mr. Akash. I have no reason to doubt his report and when I went through it, his methodology was sound. It produced the expected results. The only thing I disagree about his report is the conclusion that he drew, which is that, the moment you see the word Leaderonomics there, then you must have mean that Leaderonomics do it. That is the part that I disagree. … DW3 Okay, so, No. 1, I’ve looked at his report. I studied his methodology, looks fine, no problem. I agree with how he went about coming to his, I mean getting the answers that he saw. I just don’t agree with the conclusion. And why I say that is because No. 1, when he mentioned he detected no traces of tampering upon receipt by the person who received it. So, that means from the time they received it until the time when he took a look at it, no one touched it. DW3 He doesn’t talk about what happens before it was sent out, and before it was sent out you won’t even, you may not even need to use the word tampering, it’s just a case of modification of metadata in a very legitimate manner through modifying settings within Microsoft Word. You can do that. And in some cases, you can also do a right click of the file itself before you send it out, and change the author properties from there. There is actually a few ways to go about doing it. You don’t even have to introduce the word tampering at all. It’s not malicious. It just right there. Re-Examination of DW3, Tan Keh Win [Notes of Proceedings page 343] DHR Just a few questions only My Lord, only to clarify. Mr. Tan you were asked a series of questions about a forensic software being used to analyse the late D2’s laptop. Should you recall that and you were asked whether traces of the Poison Pen Letter of MACC.docx can be found, you said yes. For record you wanted to explain. Do you stand by your answer or do you wish to add on to it? DW3 No, it’s just the case of, if we were to look at the laptop that we think contains this file, you may or may not find the file because we are not sure if it even came from that particular laptop. In fact, if we had a, if it is eventually entirely possible that the laptop used to create this file may not have even come from Leaderonomics at all. Because all you have going on based on the MACC.docx file, is that the “author” property, and the “last modified by”, is by the screen value of Leaderonomics. But you can easily replicate that on any other laptop with any other Microsoft Word application. In fact, if I wanted to and I had access to the MACC.docx file, ok, I opened that file up, I do a select all copy based on the entire text, and then keeping that information in memory, I create a brand-new Microsoft Word file, brand new Microsoft Word file, I change the options within Microsoft Word to show that “author” is now a new name, could be Tan Keh Win, no problem. I then set my system time to reflect the same time that the MACC.docx file was created, I then paste it in, I save, I’ll get the exact the same result. [Emphasis added] [92] In particular, DW3 conducted a demonstration to show that a work laptop owned by the 1st defendant could create multiple Microsoft Word documents with different metadata properties, one that carry the value of “John Doe” in the “Author” and “Last Modified By” metadata properties and another one where the name “John Doe” was changed to “Leaderonomics” and the name “Leaderonomics” was reflected in the metadata properties of a new created Microsoft Word document. [See Forensic Report of DW3]. [93] I find Tan Keh Win/DW3’s testimony confident and his testimony was never seriously challenged by counsel for the plaintiff. In fact, his testimony was shared by PW4 Steven Wilkinson’s as was candidly admitted by the plaintiff in his Reply Submissions at paragraph 46 - … the Plaintiff respectfully submits that DW3’s evidence on the metadata and properties of the Poison Pen Letter, at best, only goes towards the possibility that in principle the metadata of a Microsoft Word document may be altered (view of which Expert Steven also shared with DW3 – page 194 of NOE). [Emphasis added] [94] After hearing the three experts, I am of the view that metadata analysis can be manipulated and cannot properly be regarded as reliable evidence supportive of either side’s case in a court dispute. [95] Decisions of courts from various jurisdictions have come to similar conclusion.[Maui Ashley Solomon v David James Prater [2021] NZHC 481 (High Court, New Zealand, Commonwealth v Davis Samuel Pty Ltd And Others (No 7) [2013] ACTSC 146 (Supreme Court of the Australian, Boyapati and Others v Rockefeller Management Corporation and Others [2008] FCA 995 (Federal Court of Australia), Croftcall Ltd v Morgan and another [2008] All ER