• Security incident: ISF was recently accessed by intruders. Please change your password, and change it anywhere else you used it. Read more

Merged Arday life and death

I was about to say I think colleges would greatly benefit from having their HR departments run a candidate's degree thesis through a plagiarism detection tool. It doesn't even have to be dispositive, just something for HR to flag up to the appointment committee for closer examination.

I was about to say that, but then I realized it would probably reflect badly on undergraduate programs everywhere, and could even possibly bring the whole edifice of higher education crashing down. So maybe the occasional loss of a single life is a small price to pay, to keep all of higher education from crumbling to dust. Arday may be dead, but Oxbridge lives on. Perhaps it's for the best.
 
Well this is pretty naive of you.

One of my daughters is an executive in one of Australia’s biggest companies and her remit includes recruitment. She has told me that people lie all the time about qualifications. Her company uses
AI to dive deeply into claims of qualifications, including plagiarism for senior jobs. It takes little time and doesn’t cost a lot.
He was employed in 2023, I doubt the AI tools that your daughter uses existed back then.
 
Much has been made of the falsehoods he produced. Much has been said, both implicitly and explicitly, about how it's racist to talk about his shortcomings as an academic.
The closest I've seen anyone here get to that accusation is 'do people care this much when anyone else is caught out in bad academic behaviour, whose qualifications don't hold up etc' but I will grant that this guy did tell some attention-grabbing porkies and it does end up looking very bad that he was never gently removed from any hastily provided pedestals until it got this bad. And Cambridge had years of performance they could have been evaluating seriously.

It is not racist to discuss this guy's shortcomings as an academic.

It can be racist to extrapolate too fabulously about what the guy's ability to skate on cool stories (until he couldn't) says about how people are doing things and what people, movements, and institutions are like. He's just one guy and dumb ◊◊◊◊ happens sometimes. If you want to use the position he rose to as a data point for the school admin processes being ◊◊◊◊, I can't complain, but he's still just one example (even if he's a perfect example).
 
Last edited:
As a bit of a tangent, my PhD supervisor was brilliant but not always in contact with reality, but she did have an excellent eye for a PhD student and got her name on lots of decent publications as a result. When I was leaving the university I had a coffee with the Dean and he remarked that it had been noticed how her publications in English had seen a major upswing in quality during my time in the department :ROFLMAO:.

Yeah, academic fraud of some kind or another is absolutely rife, even among the allegedly "harder" subjects.

A friend of mine is a computer scientist who studied at the University of Salamanca was shocked to find that papers he had written ended up with another professor's name attached (as co-author or as a citation, I don't remember). The same professor was continuously inflating his own h-index through self-citation (sometimes a normal practice, but in this case taken to excess!), and having mysterious other professors (who may not even have existed) cite him as well.

Link

In response to concerns raised by several scientists and a prominent Spanish media outlet, the scientific publisher Springer Nature has retracted 75 conference papers connected to computer scientist Juan Manuel Corchado, rector of the University of Salamanca (Usal), who has been accused of dramatically inflating citations to his own work.

The retraction notices—the last of which were published today—note “unusual citation behavior,” among other issues. The papers cite Corchado 1772 times, with a further 559 references to the journal that he edits, the Advances in Distributed Computing and Artificial Intelligence Journal (ADCAIJ), and another 329 citations to members of Corchado’s research group. Corchado authored 14 of the papers himself, but most included at least one author from Usal.

Sneaky bastard...

Corchado told seven lies in his reply to the information published by this newspaper and which he posted on his website with the title Defending the truth. The professor claimed that the documents with thousands of self-citations were simply “class exercises posted on a university website.” That’s the first lie. The reality is that Corchado used the same trick in his presentations at conferences. In a two-page abstract for a conference in Chennai, India, he cited himself 200 times. The academic knew that the Google Scholar search engine would track these documents and take them into account to develop its metrics, which is why Corchado appears to be one of the experts in artificial intelligence with the greatest impact in the world, without actually being one. Corchado has ignored new requests for information from this newspaper.
 
I was about to say I think colleges would greatly benefit from having their HR departments run a candidate's degree thesis through a plagiarism detection tool. It doesn't even have to be dispositive, just something for HR to flag up to the appointment committee for closer examination.

I was about to say that, but then I realized it would probably reflect badly on undergraduate programs everywhere, and could even possibly bring the whole edifice of higher education crashing down. So maybe the occasional loss of a single life is a small price to pay, to keep all of higher education from crumbling to dust. Arday may be dead, but Oxbridge lives on. Perhaps it's for the best.
An academic who understands how to cite correctly still needs to examine any flagged text to see if the source is acknowledged, since there really isn't any such thing as a plagiarism detection tool. It shouldn't be necessary though, as these checks because it should be done when the thesis was examined. Something is going wrong that possibly relates to universities not wanting to have student failure on their record.

And plagiarism does not require any matching to source materials at all. In the days before we had these tools, if we suspected there was unattributed material in a piece of work but we couldn't locate the source we would send it round to colleagues asking if they recognised it. Nobody cared whether the exact words of the source were used, only whether it was clearly influenced by a source that was not cited. Of course if we go back to the ancient times of the 90s when I did my PhD, most students were still using typewriters and they didn't tend to use the exact words of a source as much anyway. That started with copy and paste.
 
Multiple people in that field have come forward to say that they realised at the time that his publication record was pathetic for a job of that nature. They didn't say anything at the time of course, for fear of accusations of racism.

....snip...

For a senior professorial chair, particularly when considering an unusually young candidate, the appointments committee should be taking a very deep dive into his work
I would also expect some informal conversations with people the candidate had not named as referees but had worked alongside or under him.
...snip...
That would open the university to lawsuits, whilst of course in the real-world such conversations do happen even though they shouldn't it would be very difficult to use the results as a reason to drop a candidate. The principle in the UK is you interview the person in front of you, any questions about qualifications etc. should be made to the candidate, likewise pre-employment vetting has to be very limited, outside of legal requirements - such as right-to-work verification it has to be for very specific reasons - you can't just go and search for every candidates on social media (and being nosey doesn't count). My last two interactions with largish companies and their "HR" departments (about 8 years ago now) in regards to recruiting (new teams from senior to entry level positions) had company policies that flatly forbid the use of information from social media when deciding whether to employ anyone. One also made it a requirement to declare if you had seen anything from social media about a candidate. You usually in any reasonably medium and upwards sized company have to handover notes you made and so on during the selection process to the HR department. A well run HR department will brief interviewers as to what is appropriate (and inappropriate) to ask and what can and can't be be used as a criteria for hiring.

Cambridge could be seeing a lawsuit from Arday's family and that could cover why he was employed in the first place, if Cambridge screwed up and employed him because he was black a lack of "duty of care" case would be on a strong footing. I hope they had all their ducks in order and documented everything else this could be financially costly for them.
 
Multiple people in that field have come forward to say that they realised at the time that his publication record was pathetic for a job of that nature. They didn't say anything at the time of course, for fear of accusations of racism.d
Far be it from me to defend someone whose work may well be rubbish, but what kind of publication record do you think they would look at?

I think Jason Stanley's claim that his work was not plagiarized is nonsense, but he is probably correct that the superficial record at least was not bad.

Link

Just by way of a rough comparison, I am currently interested in reading books on the history of Central Asia, and a leading historian of that area has a professorial role at Oxford yet only has an h-index of 9, compared with Arday's 20 or 21.

Link

Of course, such publication records/h-indexes are hardly without problems, as you point out...

A strong publication record, both as regards numbers and depth, is the main thing an appointments committee will look at. That and the ability to attract research funding, and having headed up major research projects. (Teaching ability is a fair way down the list, at least as regards undergraduates. They will be expected to have supervised many PhD students through to graduation though. Arday hadn't done a single one.)
I think the point about being able to attract research funding is an important one here.

Some people have immediately jumped on the bandwagon of assuming that Arday was there because Cambridge wanted to prove they are the "wokest of the woke, the holiest of the holies etc....", but who, these days, are in charge of bringing in funding? In many cases universities are driven by the same economic interests as influencers. Who brings the eyeballs? etc...

Like it or not, if they thought Arday was a ticket to publicity and funding, then it probably makes sense that they responded to those incentives as much as anything else.
I remember when one of my PhD supervisors was promoted to Senior Lecturer. I had by then realised that he was a bit of an intellectual lightweight, and was being carried along by being able to talk a good line, and by the efforts of his numerous PhD students whose work he assimilated. He got them to write papers, which was good for their CVs of course and their academic development, but then he got his name on all the papers. Heard in the tea-room later, "Well have you seen his publications list? It ran to five pages. All they did was count the papers and give the job to the guy with the largest number."
Indeed! This is something that has been going on in academia for a long time, and again, not just in the "soft" subjects.
He wasn't useless, but I relied more on my other supervisor, he just wasn't a heavyweight. He left for another job where they saw through him and he was sacked. I later went up to his former boss at a BVA meeting and said, I just want to shake the hand of the man who sacked David Snow!

For a senior professorial chair, particularly when considering an unusually young candidate, the appointments committee should be taking a very deep dive into his work I would also expect some informal conversations with people the candidate had not named as referees but had worked alongside or under him.

By all accounts, in addition to his PhD being plagiarised "quotes" linked together by meaningless waffle, his publications were few in number and lightweight, some merely detailing a handful of conversations he'd had with people, or even with a friend. These also showed both external and internal plagiarism, copying from various other sources and then repeating the same material from paper to paper. If you listen to David Harris's video, this was what he first noticed. He was writing a book about the subject, and when it was nearly completed someone drew his attention to the existence of Jason Arday, whom he had literally never heard of. He felt this was a serious omission, and started to skim-read Arday's papers to get a handle on his research. Reading them one after another like that he spotted the same quotes being attributed to different subjects in different papers, and other anomalies and inconsistencies.

If David Harris could do it, the appointments committee could do it.
Yep, Arday should not have been given the job, no doubt. But I would have thought that such a realization would have come out even in the interview. That said, if people just tend to have the gift of the gab and are able to sidestep serious questions, it ironically diminishes the viva that we were talking about earlier.
 
That would open the university to lawsuits, whilst of course in the real-world such conversations do happen even though they shouldn't it would be very difficult to use the results as a reason to drop a candidate. The principle in the UK is you interview the person in front of you, any questions about qualifications etc. should be made to the candidate, likewise pre-employment vetting has to be very limited, outside of legal requirements - such as right-to-work verification it has to be for very specific reasons - you can't just go and search for every candidates on social media (and being nosey doesn't count). My last two interactions with largish companies and their "HR" departments (about 8 years ago now) in regards to recruiting (new teams from senior to entry level positions) had company policies that flatly forbid the use of information from social media when deciding whether to employ anyone. One also made it a requirement to declare if you had seen anything from social media about a candidate. You usually in any reasonably medium and upwards sized company have to handover notes you made and so on during the selection process to the HR department. A well run HR department will brief interviewers as to what is appropriate (and inappropriate) to ask and what can and can't be be used as a criteria for hiring.

Cambridge could be seeing a lawsuit from Arday's family and that could cover why he was employed in the first place, if Cambridge screwed up and employed him because he was black a lack of "duty of care" case would be on a strong footing. I hope they had all their ducks in order and documented everything else this could be financially costly for them.
Wow. You have no idea of how HR departments operate today and how they utilise AI.

As a side issue you also have no idea of the reliability of selection interviews. I do. But this is probably an issue for another thread, which I would be happy to join.
 
Turnitin was was an alive and functioning plagiarism program well before then. Look it up.
Yes, I have used Turnitin extensively, and I have done so since around 2016. That said, I don't think it has been common practice to feed the journal articles and PhDs of interview candidates into Turnitin. It is, instead, used for the submission of student work. The expectation would be that the awarding institution would have done their due diligence as well as the journals.

You might respond that LJMU should have used Turnitin, and that would be fair, but the issue in this case is Cambridge's job application process.
 
The reality is that Corchado used the same trick in his presentations at conferences. In a two-page abstract for a conference in Chennai, India, he cited himself 200 times. The academic knew that the Google Scholar search engine would track these documents and take them into account to develop its metrics, which is why Corchado appears to be one of the experts in artificial intelligence with the greatest impact in the world, without actually being one. Corchado has ignored new requests for information from this newspaper.
Wondering how anyone could cram 200 self-citations into a two-page abstract, I took a look. It's a two-page abstract with 16 pages of references. How could anyone pack 322 references into a two-page abstract? Here's how:
This means that the Deepint.net platform integrates functionalities that would normally take years to implement in any sector individually and that would normally require a group of experts in data analysis and related technologies [1-322] .
Sheesh.
 
Wondering how anyone could cram 200 self-citations into a two-page abstract, I took a look. It's a two-page abstract with 16 pages of references. How could anyone pack 322 references into a two-page abstract? Here's how:

Sheesh.
Yes, he is notorious. When I saw the story about him I wrote to my friend because I knew he had studied there. His response was effectively that he was a very well-known entity and had been doing this flagrantly for years.
 
Yeah, academic fraud of some kind or another is absolutely rife, even among the allegedly "harder" subjects.

A friend of mine is a computer scientist who studied at the University of Salamanca was shocked to find that papers he had written ended up with another professor's name attached (as co-author or as a citation, I don't remember). The same professor was continuously inflating his own h-index through self-citation (sometimes a normal practice, but in this case taken to excess!), and having mysterious other professors (who may not even have existed) cite him as well.

Link



Sneaky bastard...
He makes Arday look like a choir boy!
 
He makes Arday look like a choir boy!
Yep... his Wikipedia page:

In 2022, Corchado's publications were analysed by Retraction Watch, which found that about 22% of the total citation he received for his publications were from his own research papers, a serious case of excessive self-citation.[5] In 2023, a whistleblower informed Spanish newspaper El País the possible extent of publication misconduct perpetrated by Corchado. A journalistic investigation revealed evidence of several instances on self-citations. Corchado had 8,700 citations in the Web of Science and 45,000 in Google Scholar, many from his own works.
These are pretty crazy. He invented fake scholars to cite himself.
In an extreme case, he cited 100 of his own papers in a four-paragraphed article. Some authors who profusely cited his works turned out to be fictitious and were traced back to his own profile.[6]

https://en.wikipedia.org/wiki/Juan_Manuel_Corchado#cite_note-7
Springer Nature also investigated his papers they published, finding out not only incidences of self-citation but also "unusual citation behavior",
he had occasional coercive citations by influencing others, especially researchers from his university, to cite his works. In 2024, the publisher retracted 75 papers related to his papers out of which 14 were his own authorship.[7]

There you go! "Coercive ctiations" as well!
 
Wow.
You have no idea of how HR departments operate today and how they utilise AI.
As a side issue you also have no idea of the reliability of selection interviews. I do. But this is probably an issue for another thread, which I would be happy to join.
Granted not how they use LLMs today but I do know a lot about HR departments after being responsible for a few in rather large companies and I am still aware of the legal framework in which recruitment in the UK operates. Have you considered the legal framework in Australia could be different to the UK? For example here is the UK's advice on pre-vetting: https://ico.org.uk/for-organisation...lection/pre-employment-vetting-of-candidates/ Does Australia have the same GDPR regulations that the UK has and its implications on recruitment and what you can lawfully use and what you can't? i.e. https://ico.org.uk/for-organisation...nd-selection/data-protection-and-recruitment/

ETA: I should add I have no direct knowledge of recruiting for an academic position, my experience is limited to positions in the business world so what is legal and illegal in the UK in regards to all recruitment.
 
Last edited:
Yes, I have used Turnitin extensively, and I have done so since around 2016. That said, I don't think it has been common practice to feed the journal articles and PhDs of interview candidates into Turnitin. It is, instead, used for the submission of student work. The expectation would be that the awarding institution would have done their due diligence as well as the journals.

You might respond that LJMU should have used Turnitin, and that would be fair, but the issue in this case is Cambridge's job application process.
When did they start to brand it as "AI" - let me guess around 2023....
 
When did they start to brand it as "AI" - let me guess around 2023....
Good question. My guess is that they started to call it an AI detector when no student who would otherwise have plagiarized had any reason to do it any more. All such students would simply feed their essay questions into an AI and have it churn out the answer.

In my experience, Turnitin is pure garbage as an AI detector, and I don't believe its claims as having something like a 1% false positive/negative result. It's probably worse than human intuition.

On the other hand, Pangram seems to be genuinely better.

By the way, there is another wrinkle with the self-plagiarism angle. It seems that some people have noticed that Cofnas has "self-plagiarized" his PhD thesis, yet this is also a fairly normal process in that many PhD theses will be composed of, or subsequently published as, a number of papers.

Someone I know also had some trouble when either a pre-print of theirs or maybe a pre-registration of their methods flagged their paper as plagiarized.
 
That would open the university to lawsuits, whilst of course in the real-world such conversations do happen even though they shouldn't it would be very difficult to use the results as a reason to drop a candidate. The principle in the UK is you interview the person in front of you, any questions about qualifications etc. should be made to the candidate, likewise pre-employment vetting has to be very limited, outside of legal requirements - such as right-to-work verification it has to be for very specific reasons - you can't just go and search for every candidates on social media (and being nosey doesn't count). My last two interactions with largish companies and their "HR" departments (about 8 years ago now) in regards to recruiting (new teams from senior to entry level positions) had company policies that flatly forbid the use of information from social media when deciding whether to employ anyone. One also made it a requirement to declare if you had seen anything from social media about a candidate. You usually in any reasonably medium and upwards sized company have to handover notes you made and so on during the selection process to the HR department. A well run HR department will brief interviewers as to what is appropriate (and inappropriate) to ask and what can and can't be be used as a criteria for hiring.

Cambridge could be seeing a lawsuit from Arday's family and that could cover why he was employed in the first place, if Cambridge screwed up and employed him because he was black a lack of "duty of care" case would be on a strong footing. I hope they had all their ducks in order and documented everything else this could be financially costly for them.

And yet, people who were privy to the appointment said that it looked as if the committee were just going through the motions, having already decided to appoint Arday. I wonder how they came to that position.

I'm not imagining any informal conversations would be on the record. The point is, informal. You test the water and listen to a variety of opinions on the candidate from people who have no idea why you might be interested, and you form a view. That's it. This guy was a chancer, and there were people who knew it. On the other hand there's more than a slight suspicion that people in the establishment he was leaving might have been very disinclined to say anything that might obstruct the departure process.
 
Turnitin was was an alive and functioning plagiarism program well before then. Look it up.

I retired in 2018 and I was using Turnitin for the last few years of my teaching career. I didn't use the plagiarism feature though.
 
Wondering how anyone could cram 200 self-citations into a two-page abstract, I took a look. It's a two-page abstract with 16 pages of references. How could anyone pack 322 references into a two-page abstract? Here's how:

Sheesh.

Bloody hell. Surely that would raise a few suspicions. Or at least eyebrows.
 

ISF - Join now!

Every member here is approved by hand. No bots, no spam, just people who care about evidence and honest debate.

Membership is free!

Create your free account

Back
Top Bottom