Was it Slander?
This is a historical summary of a controversy between me and Professor Mark Orazem of the University of Florida and his associates. Its presence here has been made necessary by the failure of these authors to respond to repeated requests for clarification or withdrawal of their charges against me (see the end of Item E listed below and the copy of an email message at the end of this summary). The controversy first involved the charge by Orazem and his associate, Professor L. Garcia-Rubio, (acronym: OGR) that I and my Russian associate, Professor Vladimir Piterbarg, had slandered them, and it has escalated from there! I believe that the present material shows that all of OGR’s charges are incorrect and that they have slandered me. Since all the relevant material involved is included here, the reader may judge for him or herself. It is a sad saga, but the reader may find it interesting and instructive to see the unfortunate consequences (both for me and for OGR) of my doing something for what I considered the best of motives: to help another scientist and to advance science.
The interested reader may find it useful to begin by reading the essay “What’s wrong with these publications,” included in the present web site. Although it was written by me some years before any of the present problems arose, some points discussed in it are relevant to these problems and to what led to them.
I (acronym: JRM) suggest that the reader complete a reading of this summary and then fill in the details by reading all or any of the here-appended copies of the correspondence relevant to the matter. They are listed below in temporal order, and one can go directly from any of these items to the actual material by clicking on the listing letter.
- Letter from OGR to Roger Parsons, editor of J. Electroanal. Chem., 2/19/98
- Email memo from JRM to Mark Orazem, 4/16/98
- Letter from OGR to JRM, 5/4/98
- Letter from JRM to OGR, 5/26/98
- Letter from JRM to OGR, 6/12/98.
The history of this affair begins with my paper in Electrochim. Acta 38(1993)1883, one in which I asked the question of how random errors in small-signal frequency-response data would transform under the Kramers-Kronig [KK] integral transformations and answered it by means of a Monte Carlo study. At that time, I and my associate, Professor William J. Thompson, considered deriving an analytical proof of the Monte Carlo results but were diverted by more pressing matters. So things remained until the spring of 1996. Over the years, I have provided several long and detailed reviews of impedance-spectroscopy-related manuscripts by Orazem and his associates (all of which I recommended for publication after some changes), so I was not surprised to receive one from the Journal of the Electrochemical Society [JECS] in late March 1996 written by M. Durbha (an Orazem graduate student), Orazem, and Garcia-Rubio [DOGR], entitled, “Error Analysis for Spectroscopy Applications of the Kramers-Kronig Transforms: Electrochemical Impedance Spectroscopy.”
Unfortunately, I found that this MS contained many problems and errors, and so I recommended in my detailed review that it only be published after all important errors were corrected. Although the MS contained an analytic derivation of the transformation of stochastic errors under the KK transforms, this proof was faulty and completely inapplicable. As I stated in the review, it was “fatally flawed” and even involved two equations containing elementary errors: ones in which a real integral was set equal to an imaginary quantity (Eqs. 16 and B-5), rather than to the correct results of zero or a delta function, respectively! At the time, I did not know how to provide a complete correct proof, so no path toward one was suggested in the review.
The review was sent to the requesting editor by email on 4/16/96. Working on it had stimulated me to think again, however, about the need to develop an adequate proof, so I discussed the matter with Professor Piterbarg, a distinguished visiting statistician. Since it seemed highly unlikely that DOGR would produce such a proof, especially since they were evidently unaware of the total inadequacy of their own approach and since the matter involved a problem I had originated, we felt it proper and reasonable for us to try to provide a correct analytical proof, thereby advancing science. We (largely Piterbarg) were successful in doing so, and we wrote a short manuscript (acronym: MP) on the subject. At the suggestion of the JECS editor to whom my review was sent, it was submitted to JECS, and we received an acknowledgment of its receipt dated 6/28/1996.
At the same time that this MS was submitted to the journal (6/23/96), a copy of it and a covering letter were sent to Mark Orazem. In this letter, I identified myself as a reviewer of the DOGR MS and invited comments and suggestions about our MS. Although I said in the letter that I was sorry that I could not recommend the DOGR MS for publication, in the review of the DOGR MS I concluded that “if the authors believe that they have … (produced a corrected proof) …, a revised MS should be submitted.” Thus, I did not advise rejection of the MS, contrary to the OGR assertion in item A. I received no response from Orazem to my letter and none to subsequent email messages.
Although it would have been reasonable for me to have been asked to review any revised version of the DOGR MS, this did not happen, and I received a letter dated 9/20/96 from the top JECS editor, Paul Kohl, thanking me for the DOGR review and for one of another MS. In this letter Kohl stated that the authors of the DOGR MS have revised their MS and corrected the errors, rendering the MS acceptable for publication. Later comparison with the published version of their MS showed that, in fact, they made very few of the corrections suggested in my review of the original MS. In their published version, the date of receipt of their revised (and accepted) MS is listed as 9/27/96.
On 9/23/96, I received a letter from Kohl stating that although the MP MS contained publishable material, it could not be accepted for publication in JECS in its current form because it involved a “more esoteric subject” and lacked sufficient impact for the readers of the Journal. It is reasonable to assume that Mark Orazem may have been one of the reviewers of this MS. Piterbarg and I decided not to revise it but instead submitted it in its original form to J. Electroanal. Chem.; it was later accepted with only a few grammatical corrections.
The revised version of the DOGR MS was published in JECS in the January 1997 issue, p. 48. When Piterbarg and I became aware of it, we were dismayed to find that although it now contained a KK transformation proof entirely different from that in the original version of the MS and one remarkably similar to the one present in the MP MS, no mention of our prior work was included. It seemed to us, therefore, that we were forced to act to protect the priority of our work since the DOGR paper appeared several months earlier than the MP one could appear. We therefore added a note-added-in-proof [NAIP] to our work, one which stated the main facts of what had happened concerning the MP MS. See Item B in the above listing for more details. But before submitting the proof for publication, I called Luis Garcia-Rubio, discussed the problem with him and told him about the NAIP and the main points in it. In an effort to avoid the necessity of publishing it, I asked him to discuss the matter with Orazem and provide suggestions of how we might either revise the NAIP or even eliminate it if OGR would suggest a satisfactory alternative. Unfortunately, they did not respond to this request and never even asked for a copy of the NAIP, so we reluctantly included the NAIP in the proof. A copy of it is included at the end of this summary.
Several months after the appearance of the MP paper containing the NAIP, OGR wrote an intemperate letter (Item A) to Roger Parsons. They challenged the accuracy of our recital of the facts of the situation and stated that the NAIP was a one-sided, slanderous attack on their reputations. They also erroneously stated that they had had no opportunity to rebut the claims we made in the NAIP. In view of my discussion with Garcia-Rubio described above, this statement seems particularly disingenuous.
We took the OGR charge of slander seriously, even though we were confident that we had not misrepresented any aspects of the situation in the NAIP. Therefore, in Item B I wrote a long response to the OGR assertions of Item A and requested that OGR furnish us with all examples of where we may have incorrectly stated the facts of the matter. I offered to publish a retraction and correction of any such errors. This request was reiterated in Items D and E. At the present time of writing, several months after the Item-B response was sent, OGR have not responded to this request. Thus, we conclude that they have been unable to document any such errors, and their charge of slander is incorrect.
Also in Item A, OGR charged me with several other scientific misdeeds and repeated some of them in Item C. Although these matters were not relevant to the above charge of slander, they may have been included in order to attempt to impugn my “personal and professional integrity” (quoted from Item C). Clearly, such charges would not only draw attention away from the matters discussed in the NAIP but would also, if true, tend to reduce or destroy the reader’s belief in the truth of the statements in the NAIP. Some of these charges have been discussed in Item B, and the rest of them in Items D and E. There, I believe that they have been shown to be baseless and incorrect. If so, the OGR failure to document their charge of slander and their other charges against me may themselves be considered slanderous.
In Item A, OGR state in the NAIP that I (not I and Piterbarg!) imply that OGR plagiarized the proof in our MP MS. We never mentioned the word plagiarism but agree that one interpretation of the facts stated in the NAIP would be that plagiarism took place. We stand only by the facts; their interpretation is a matter for the individual reader. It is worthy of note that although OGR state in Item A that there was no such plagiarism, their letter was not signed by the principal author of the DOGR paper, Mr. M. Durbha, an Orazem graduate student.
To show that there was indeed no plagiarism, it would be very helpful if the DOGR authors would discuss in detail how and why their proof contained so many common elements with ours. A possible, entirely hypothetical, explanation might be: An inexperienced graduate student, who had seen our MS, unwittingly incorporated elements of our proof in constructing that in the DOGR paper, and the senior authors of the paper did not notice such correspondence when they read and revised the paper before its final submission. This scenario is made somewhat more plausible by the presence in the original DOGR MS of elementary mathematical errors which such senior and experienced authors would be expected to find and eliminate if they had read the MS sufficiently carefully.
Finally, in Item A, OGR requested an apology from Roger Parsons for his accepting our note without providing an opportunity for them to rebut our claims (but see the above discussion of this matter and that in Item B). He has written an editorial which does so and is currently awaiting publication. In it, the NAIP is implicitly characterized as polemical and as “solely critical of another author,” and “as including no new scientific material.” As already discussed, the inclusion of the NAIP was not meant to be polemical and to involve controversial elements. Since its facts have not been shown to be incorrect by OGR, there seems to us to be no basis for their charge of slander. Further, the NAIP was not meant to be critical of DOGR but was included to establish the priority of our KK noise transformation proof. We feel that the statements about this and about the similarity of our proof and that in the DOGR paper are “new scientific material” and not matters of opinion.
Nevertheless, we wish that Parsons had indeed requested a response from OGR to be published along with the NAIP. This would have saved much subsequent time for all concerned, time better spent on science than on controversy. Even more, we wish that we had originally submitted the MP MS to another journal rather than to the JECS and had not sent a copy of it to Orazem! Piterbarg and I believe that we have been entirely truthful in all our discussions of this regrettable affair. We see any problems for DOGR arising from it as associated with their own actions and inaction at crucial junctures. The reader is invited to form his or her own conclusions.
I made a final try to avoid the necessity of going public with the present matters by sending the email message below. Unfortunately, no response to it or to my other letters has been received.
—————————————————————————————–
From Sat Jul 11 08:45:16 1998
Date: Fri, 10 Jul 1998 09:52:29 -0400 (EDT)
From: “J. Ross Macdonald” <>
To: Luis Garcia-Rubio<>
Cc: Mark Orazem <>,
Ross Macdonald <>
Subject: Controversy
Luis:
I have been trying to reach you by telephone for several days without success. Especially if you and Mark Orazem do not plan to respond to my last two letters by 17 July as I requested or if you plan to reject my request that you support your charges or withdraw them, we need to discuss the situation by phone. I do not wish to be forced to take the next step in our controversy, one which would likely escalate the matter to various outside agencies and to the electrochemical fraternity in general.
Perhaps by talking we can reach some measure of agreement. I think it is worth a try, and I hope you or Mark will call me before the 18th. of July at 919/967-5005.
Sincerely,
Ross Macdonald
Note added in proof (NAIP)
(in Macdonald and Piterbarg, J. Electroanal. Chem. 428(1997)6)
Since this paper was accepted, one which deals in part with the same subject has recently appeared (M. Durbha, M.E. Orazem, and L. H. Garcia-Rubio, J. Electrochem. Soc. 144 (1997) 48). One of the present authors (JRM) was a reviewer in April 1996 of the original version of this paper, provided a long, detailed review, and recommended that the MS be accepted only after extensive revision because its analytical proof of KKR noise-transformation was incorrect. But the Orazem work prompted the present authors to subsequently develop a completely different and correct proof. The present paper is the result, and a preprint of it (identical to the present accepted version) was sent to Dr. Orazem in June 1996, along with the identification of JRM as one of the reviewers of the original version.
The new proof in the revised Orazem paper, which was received by the journal near the end of September 1996 but not seen by the present authors until its publication, is quite different from their original one and is less general and more approximate than that presented here. It is noteworthy that in the Orazem paper neither are the referees thanked for their comments and suggestions nor is the early receipt of, or existence of, the present work acknowledged. Nevertheless, on comparing the final version of the Orazem paper with this one, it is obvious that their proof involves many crucial elements first introduced in the present work. In particular: (a) the basic idea of proving the equality of the input and output error variances using the rules and constraints of numerical integration; (b) the essence of their Eqs. 13, 14, 15, 27, and 28; and (c) the recognition that output stochastic errors are generated in the immediate neighborhood of the pole in the integrand. Their proof does not contain specific definitions of the placement of their integration points, ym, or of their weights, yet the present work shows that the transformed variances depend sensitively on such details as the specific form of the numerical integration procedure. Thus their proof is both derivative and sufficiently incomplete to preclude its application to actual numerical integration without further information. The Orazem work introduces a Taylor expansion, stated to be valid when the error variance is continuous at the pole position. However, such an expansion requires the existence of two-times differentiability (with a continuous second derivative), conditions not mentioned. Further, all the proofs in the present work require no differentiability conditions, are valid for non-constant but continuous variance, and do not require the assumption of a particular error structure.
Finally, the Orazem paper contains incorrect and inappropriate criticisms of the original JRM Monte-Carlo work (Ref. 6 herein), ones absent from their original version, but perhaps related to the recommendation in the JRM review that their actual references to the earlier work were misleading and misplaced and that the results of the earlier work should be recognized as the justification for developing an analytic proof of the noise transformation relations. After reading the above discussion of the history of these matters, the reader will not be surprised to learn that this recommendation was not implemented.
————————————————————————————
UNIVERSITY OF FLORIDA
College of Engineering
PO Box 116005
Department of Chemical Engineering
Gainesville, FL 32611-6005
Phone: (352) 392-0881
February 19, 1998
Roger Parsons
Editor, Journal of Electroanalytical Chemistry
University of Southhampton
Department of Chemistry
Southhampton, S017 1BJ
United Kingdom
Re: Macdonald and Piterbarg, Journal of Electroanalytical Chemistry, 428 (1997), 1-9.
Dear Professor Parsons:
We have read the above referenced article by Macdonald and Piterbarg in which Professor Macdonald added a note-in-proof that attacks our integrity and that of the editorial board of the Journal of the Electrochemical Society. We have read as well the correspondence between you and Paul Kohl, the editor of the Journal of the Electrochemical Society. We understand, from your correspondence, that you were fully aware of the content and the personal nature of the note added in proof and that you did not act to verify the claims made. You should be aware that his comments do not describe properly the facts of the matter and, therefore, constitute slander. This is a serious issue, and, as no opportunity to rebut the claims has been provided, the Journal of Electroanalytical Chemistry and the parent company Elsevier are responsible for publishing a one-sided attack on our reputation.
The note-added-in-proof centers about Professor Macdonald’s review of a manuscript we submitted to the Journal of the Electrochemical Society. Professor Macdonald identified an error in our derivation, wrote a review advising that the manuscript be rejected, and submitted to the journal a corrected version of our proof. His manuscript was submitted to the journal even before we had received the reviewers’ comments from the division editor. It seems to us that Professor Macdonald took advantage of the delay in communications to submit his paper before we could address the reviewers’ comments. Professor Macdonald’s actions constitute a serious violation of the integrity of the review process. His subsequent insertion of a personal attack into your journal must be taken in the context of this and previous actions with regards to our work.
In 1994, Professor Macdonald published a paper with B. A. Boukamp (Solid State Ionics, 74 (1994), 85) in which he claimed credit for the concept of using distributed-relaxation-time models to assess consistency with the Kramers-Kronig relations. His sole reference to our related effort reads:
Although the use of the top circuit (without the CP, CD, and RD elements) has been recently usefully advocated for preliminary “measurement model” data fitting [3 1,32], these DRT or DAE circuits have been used for fitting for years.
In this paper, Professor Macdonald took credit for our development of a new way to assess consistency with the Kramers-Kronig relations. We had been in communication with Professor Macdonald since 1992 concerning our approach. In fact, he was invited to participate with us in our developments during his visit to tile University of Florida on March 29, 1993. For example, in a letter dated June 16, 1992, Professor Macdonald acknowledged receipt of our manuscripts and credited us with development of the new concept:
“Thanks very much for the preprints/reprints you recently sent me … Here, I will talk primarily about your new KK approach as exemplified in your March 10, 1992 ASTM paper/preprint. I found your approach ingenious, and it will undoubtedly be useful … Your new approach seems to me to consist of combining two old, well-known ideas to produce a useful new technique. The first idea is that any reasonable equivalent circuit impedance will automatically satisfy the KK relations. The second idea is that of using a distribution of relaxation times circuit as the fitting circuit…”
Our frequent letter and e-mail correspondence from Professor Macdonald shows how our work motivated his development of a measurement model module and new weighting strategies in his regression software. Yet, in his 1994 paper, Professor Macdonald failed to credit us for the use of measurement models to assess consistency with the Kramers-Kronig relations. We received an acceptable apology from Professor Boukamp who stated he was unaware of our work in this area. We did not receive an acceptable apology from Professor Macdonald. When asked to explain his position, Professor Macdonald stated that he was justified because we had sent him only prepublication versions of the manuscripts.
We turn now to the Macdonald and Piterbarg manuscript published in your journal:
1. By not citing our 1991-93 archival publications on the use of measurement models for
evaluating consistency with the Kramers-Kronig relations, Macdonald and Piterbarg give the
false impression that Macdonald and Boukamp (1994) were the first to present this approach.
As stated above, Professor Macdonald received pre-prints of these papers in 1992, and he was reminded of the existence of these papers after his 1994 paper with Professor Boukamp was published.
2. In his note-added-in-proof, Professor Macdonald implies that we have plagiarized his work, complains that we did not refer to his document or acknowledge his review, and complains that our paper contained incorrect and inappropriate criticisms of his paper.
a) There was no plagiarism of his work. Once we found the error in our derivation, it was apparent that numerical integration was needed. We did not take the numerical technique from his paper. We have been employing numerical integration in our Monte Carlo simulations long before.
b) Reviewing manuscripts is an accepted professional responsibility that requires ethical behavior. It is the reviewer’s job to identify errors in manuscripts and to convey that information to the editor. Submission by the reviewer of corrected versions of manuscripts under his review is inappropriate behavior that violates the integrity of the review process. We saw no obligation on our part to thank the reviewer for his actions.
c) Our criticisms of his 1993 paper are valid and apply as well to the present paper. Professor Macdonald’s analysis of the transformation of errors through the Kramers-Kronig relations is incomplete because he has not identified correctly the conditions under which the variances are equal for real and imaginary parts of a spectrum. We have identified these conditions and have verified them experimentally for a large number of electrochemical and non-electrochemical spectroscopy measurements.
d) Within two days of receipt of the reviews of our manuscript, we received from Professor Macdonald a copy of his manuscript and a cover letter stating that it had been submitted to the Journal of the Electrochemical Society. Our receipt of this manuscript created a
few problems for us. One was that it alerted us to the fact that we were victims of unethical behavior on the part of our reviewer. Yet, the manuscript was much narrower in scope, reflecting the authors’ lack of understanding of the implications of our experimental results. Given the manuscript’s provenance as a corrected version of our paper, we did not feel that we were under obligation to describe the manuscript as the source of our ideas. It was not. In addition, we felt we could not refer to the manuscript without discussing its shortcomings and errors. Yet, we did not feel comfortable criticizing the errors because a manuscript under review is a privileged document and those errors are the responsibility of the reviewers. Since we had already established numerically the validity of our statements, we chose to work independently of his manuscript and to use the reviewers’ comments to guide correction and modification of our manuscript. We made proper reference to his 1993 paper for a variable transformation that we found to be useful.
We have made every effort to give Professor Macdonald due credit for his contributions to the field. We have more than treated him fairly, both in making him aware of our concepts and in making sure that his work is referenced properly in our manuscripts. It is evident that our good will and effort has not been reciprocated. In his paper with Boukamp, Professor Macdonald claimed credit for developments that were ours. Through his paper with Piterbarg, he violated the integrity of the review process. Under these circumstances, the note-added-in-proof constitutes an unwarranted attack on our reputations.
We are certain that, had you been aware of all the facts, you would not have agreed to publish Professor Macdonald’s note. We seek your assistance in mitigating the damage done by providing an apology or the opportunity to respond.
We would be happy to provide you copies of correspondence and other documents that support the statements made in this letter. We await your response as we decide on an appropriate course of action.
Sincerely,
(Signed by)
Mark E. Orazem, Professor
(Signed by)
Luis Garcia-Rubio, Professor
**********************************************
J. Ross Macdonald
354 Carolina Meadows, Villa
Chapel Hill, NC 27514
April 16, 1998
To: Mark Orazem, University of Florida, Gainesville
From: Ross Macdonald, University of North Carolina, Chapel Hill
Mark, I finally received a copy of the letter by you and Garcia-Rubio (OGR) addressed to Roger Parsons and dated 2/19/98. You sent my copy to my old address and it had to be forwarded on to me.
There are many points in your letter with which I disagree, but here I only want to address the main one: your accusation of slander by myself and Piterbarg. As justification, you state that the “note added in proof” (NAIP) in our paper J. Electroanal. Chem. 428(1997)1 (referred to below as MP) does not properly describe the facts in the matter. I have carefully examined both your letter and the NAIP, and I am unable to find in your letter any specifics to support your charge.
I am unaware of any later correspondence you may have had with Roger Parsons after he received your letter, but I would like to try to get to the bottom of this particular matter so that it is clearer to both of us where we agree and disagree about the “facts” you mention. If indeed it turns out that we have published incorrect facts about the matter, I wish to know that so that we can redress the balance, by publishing a retraction and correction as appropriate. Therefore, please send me a list of those factual statements that appear in the NAIP which you believe to be incorrect, along with the reasons for such belief.
Although I can understand why the OGR letter cited many earlier misdeeds alleged to have been committed by me, even were they correct, they are irrelevant to the matter of published slander, something that depends only on proving that we purposefully misrepresented the facts in the situation in our published NAIP. In providing such a list, I suggest that you stick to specific statements we published about what actually happened and perhaps about the similarities between our MS and your final one, omitting those of ours that are obviously statements of opinion only. Also, it would be helpful if you would list the errors you mention that you believe to be present in MP, those you refer to without details on p. 3 of the OGR letter.
I will note here just one place in your letter that is in variance with the facts. In the NAIP we mentioned that the review of your original MS stated that the MS only be accepted after extensive revision because its proof of the KK error transformations was incorrect. OGR have stated, however, that I wrote a review advising that the MS be rejected. The actual wording of the review states that if errors in the MS should be corrected and the points raised in the review taken into account, a revised MS should be submitted. I believe, therefore, that the NAIP statement is correct and yours is a misrepresentation of the situation.
It may be of interest to you to know that the JRM review of the original version of your MS was sent off to the editor by email on 16 April 1996. I am sorry to hear that you only received it about two months later, but I cannot be properly blamed for such delay or for not knowing about that delay.
One final point: OGR claims in several places that the MP work was a corrected version of your original MS. This claim could not be farther from the truth. As you know, in my original Monte Carlo paper of 1993 (Ref. 6 in MP), I was the first to raise the question of how errors would transform under KK, and I solved the problem numerically by Monte Carlo. It was natural for me and my colleague Professor William Thompson to think about the possibility of providing an analytical or analytic-numerical proof of the results. We considered that possibility soon after the paper was published and even talked to a professor in the Statistics Department here about the matter. Unfortunately, no solution was immediately obvious, and we postponed further work on the question because of the press of other commitments.
Any unbiased observer with appropriate background could examine your original MS and compare it with that of MP. Luckily, that is still possible if needed. Some time after I completed and submitted the review of your MS, I showed the MS to both Piterbarg, a top-notch Russian statistician, and William Thompson, a theoretical physicist and expert on numerical calculations and analysis. They both wholeheartedly agreed with me that the proof in your MS was not only incorrect but was not really even a proof. In that regard, OGR states that “once we had found the error in our derivation …” But the development of a new correct proof required not just finding an error in the original derivation, but scrapping it entirely and replacing it with an entirely new and different one.
Since the MP work developed a proof entirely different from the work in your original MS; since at the time of writing the review I was not clear about what was needed for a correct proof and did not suggest in the review any specifics for you to follow in providing one; since it seemed very likely, judging from your proposed proof, that your group was unlikely to develop an adequate one yourselves without outside help; and since I originated the field and earlier considered the need for a proof, it should be clear to you that the only thing your MS did was to stimulate Piterbarg and me to try to develop a correct proof since you had failed in doing so. Thank you for this stimulation.
We did develop a new proof, one entirely different from that in your original MS, but one that turned out to be very similar in several features to the proof appearing in the final published version of your MS, after you had had a copy of the MP work for about three months. I believe that these are all facts, ones which can be verified by an outside observer. The MP work was not a corrected version of your original MS proof work; there was nothing there worthy of copying even if we had wished to do so.
I find it surprising that you now so strongly object to the NAIP, because I believe we gave you every possible opportunity to avoid the situation we now find ourselves in. When I first sent you the MP MS on 6/23/96, I asked you for comments on the general situation, about details of the MS and questions of whether you felt we had given you proper credit for your earlier work. I said that if there were any problems from your perspective, I would rectify the matter. You did not answer this or later communications I sent you. I now see that I was grievously wrong in assuming that you would welcome a scientific advance at the possible expense of some amour-propre. I apologize for this unwarranted assumption, but I absolutely reject the OGR assertion that I committed a violation of the integrity of the review process. Had I done so, I would certainly not have identified myself to you as the reviewer of your work, sent you a copy of the MP manuscript, submitted it to the same journal as your MS, and attempted to work with you on these matters (see below).
When I first became aware in early 1997 of the revised, published version of your MS, the one with a proof very similar to that in MP, the MP paper had not yet appeared, and it was still possible to add a NAIP to it. The principal reason we decided to do so was to protect our priority in development of the proof by citing the background of the situation. Had your paper contained a proof with little or no commonality with that of MP, such as the entirely analytic one recently published by Thompson which built on the MP work, there would have been no reason to do so. Or even with the proof your paper contained, had you acknowledged in your paper that you had the MP MS during your revision, again there would have been no need for the NAIP.
But there were other opportunities you missed as well. After we put together a draft of the NAIP but while it could still be aborted, I instigated two telephone conversations in the spring of 1997 with Garcia-Rubio (not with you since you had unilaterally broken off communication with me). In these conversations, I told him what the NAIP contained and asked him to discuss with you ways which would be acceptable to all of us that would allow the NAIP to be revised or not to be published. I never had any further reply from him about this, and so the NAIP was not aborted or revised. I believe I did everything possible to try to avoid the necessity for publishing the NAIP, and that you and your co-authors evidently did not take the matter seriously and were unwilling to cooperate concerning the situation.
Finally, in the OGR letter you characterize the NAIP as a personal attack on your integrity and reputation. In our judgment, it was not personal at all, but primarily only a recital of the facts of the matter. Although the OGR letter is itself essentially a very personal attack on me specifically, by contrast the NAIP does not accuse OGR or you of any scientific or other misdeeds, such as plagiarism. That is a word you introduced in OGR. We have no way of knowing if plagiarism by OGR or by any one of the three of you took place, but we believe in giving everyone the benefit of the doubt. I do note that since you say in the OGR letter that after you received the copy of the MP MS you noted errors in it that influenced your subsequent actions and response, it is clear that some or all of you read the MS before submitting your revised MS in September 1996.
I am willing to accept your word that neither you nor Dhurba nor Garcia-Rubio used any of the ideas in the MP MS in constructing the entirely new and different proof present in the revised version of your MS. I suggest that anyone who is unwilling to do so is not a person you would want as a friend or colleague. If you now wish to affirm the absence of plagiarism by any of the three of you, that is fine, but it is still hard for me to see why you were unwilling to discuss possible resolutions of the problem with me while the revision of your MS was proceeding or why you and Garcia-Rubio were unwilling to do so when and after I talked with him, as mentioned above. It seems to me that this is what has led us to the present unfortunate situation, one for which I reject the responsibility you seem to accuse me of.
I look forward to receiving from you answers to the questions I raised above.
***************************************************
UNIVERSITY OF FLORIDA
College of Engineering
PO Box 116005
Department of Chemical Engineering
Gainesville, FL 32611-6005
May 4, 1998
J. Ross Macdonald
354 Carolina Meadows, Villa
Chapel Hill, NC 27514
Dear Ross:
Following receipt of your e-mail, we perceive a need to clarify our relationship with you.
We have great respect for your technical abilities and for your contributions to the field. We cannot, however, respect your actions toward us. When you published your paper with Bernard Boukamp in which you claimed credit for our idea of using distributed time constant models to assess consistency with the Kramers-Kronig relations, you damaged our personal and professional relationship.
You apparently do not remember the extensive communications we had with you or that we shared with you our un-published manuscripts and data. I enclose a copy of one of your communications with us in which you thank us for sending pre-prints to you and in which you recognize the novelty of our approach. This letter is dated in 1992; whereas, your manuscript was published in 1994. Your actions represent a violation of the trust we placed in you. We now feel that we can not trust in your sense of personal and professional integrity.
We see your actions during the review of our manuscript and your publication of the note-added-in-proof as being consistent with this assessment.
Sincerely,
(Signed by)
Mark E. Orazem, Professor
(Signed by)
Luis Garcia-Rubio, Professor
******************************************************************************************
J. Ross Macdonald
354 Carolina Meadows, Villa
Chapel Hill, NC 27514
May 26, 1998
Professor Mark E. Orazem
Department of Chemical Engineering
University of Florida
Gainesville, FL 32611-6005
Professor Luis H. Garcia-Rubio
Department of Chemical Engineering
University of South Florida
Tampa, FL 33620
Dear Sirs:
This is a response to your joint letter to me of 5/4/1998. In that letter you said, “When you published your paper with Bernard Boukamp in which you claimed credit for our idea of using distributed time constant models to assess consistency with the Kronig-Kramers relations, you damaged our personal and professional relationship.” I am sorry that you have this perception, and I would like to clarify the matter here. Let the acronym OGR stand for the letter of 2/19/98 which you sent to Professor Roger Parsons, one which severely criticized me and Piterbarg and alleged slander by us of your work. In that letter, you also charged me with improperly claiming the above credit.
First, thank you for sending me a copy of my letter to you dated 6/16/92. I did not have a copy of this letter in my files, and when I wrote my part of the Boukamp-Macdonald [BM] paper in 1994, I had entirely forgotten its existence and that of the three preprints/reprints you sent me in 1992 mentioned therein. Although we did not therefore reference these particular works, we did reference two of your papers relevant to the assessment of the Kronig-Kramers consistency of data by the measurement model approach (our Refs. 31 (1992) and 32 (1993)). Both of these papers of yours mention this use of the measurement model approach. It therefore seems to me that it is incorrect for you to say that I claimed credit for this application in the BM paper. Nowhere in the paper do we explicitly claim such credit, and we refer to earlier papers of yours which discuss the application. I believe that it would be proper to say that it is unfortunate that we did not refer to all of your relevant earlier work in the field and did not tie it in explicitly with the work in our paper, and I have apologized in 1995 for not doing so (see below) and I do so again here, even though I believe the matter is actually moot, as described later in this letter.
In your email to me dated 6/16/95, you first brought up the matter of our citation of your work in the above BM paper. There you characterized our citation as “incomplete and misleading,” not the same as accusing us of claiming credit for your idea. In my email reply, dated 6/19/95, having forgotten receiving the earlier preprints/reprints you cited in your email, I said that I very much regretted the omission of references to these works, and went on to say that I was unfamiliar with them (I did not say there, as you have charged in OGR, that I stated that I was justified in not referring to them because they were all pre-publication versions. If you have any evidence that I did say this, please send it to me, otherwise please withdraw this charge in your next written communication to me). Clearly, my failure to remember the earlier material you sent me was why I said that I was unaware of their existence. I much regret this lapse of memory, and I apologize for this erroneous claim in my 1995 email reply to you.
Although Bernard Boukamp was listed as the first author of BM, you have charged me alone for the above problems. Perhaps you have done so because in OGR you say that you received an acceptable apology from him but not from me. Bernard has recently written me that he believes that he never sent you a written apology but did so orally at a conference where you were both present. I believe that, except for the above lapse of memory, my written apology sent to you on 6/19/95 should also have been considered acceptable. Had you judged otherwise and felt then that I improperly claimed credit for your work and weaseled out by characterizing the earlier uncited references as unpublished, you could have raised these issues then with me. Instead, you did not do so, and I have in my files, among others, subsequent letters and emails from Mark Orazem to me dated 7/28/95, 8/2/95, 8/24/95, 10/24/95, 10/26/95, 2/26/96, and 3/8/96 which do not mention any of these matters and which are completely cordial.
If indeed you felt in 1995 or early 1996 that any of my actions damaged our personal and professional relationships, why did you not mention this to me then, after receiving my 6/19/95 apology, and ask for further explanation instead of carrying on with much cordial subsequent correspondence? I can only assume that the difference between then and now is that now you need a stick to beat me with, as you do in the OGR letter, while then you were satisfied with my apology.
May I next refer you to my letter to you of 6/16/92, which you recently reminded me of. In that letter, I pointed out to you that your measurement model approach for assessing KK consistency (particularly that first described in your 1992 ASTM preprint, published in 1993) itself consisted of the combination of two old ideas. Although I there characterized the novel element in your work as the combination of these old, well known ideas, I want to point out that the essence of the measurement model approach, and its application to KK assessment and consistency are explicitly described on p. 181 in a chapter by me appearing in the book Impedance Spectroscopy, Wiley, 1987, several years before you and your various associates invented the nomenclature “measurement model” and subsequently applied it. The only significant difference between this 1987 description and your later work is that in 1987 I suggested the use of a “practical fitting model” without explicitly citing the Voigt model you have used. But, of course, any adequate fitting model can be used – the Voigt model is only one possibility. Thus, the earlier description is the more general.
It is scarcely necessary to point out to you that I have never “claimed credit” to you in private or in publications for my above idea or even characterized your lack of referring to my prior work in this field as “incomplete and misleading.” The matter did not then seem important enough to fight about. It is still, to me, not worth an argument, but in view of your overt ad hominem disparagement, in your recent 5/4/1998 letter to me, of my “personal and professional integrity,” evidently because I gave you less credit than you felt you deserved, as discussed above, I must point out the apparent partial parallel between my inadvertent failure to credit you fully and your own complete failure to credit me for the essence of the idea which you now claim as wholly yours, one which was actually first described earlier by me rather than by you. In view of all this, it scarcely seems to matter much that we gave you less credit in BM than you would have liked for an idea and approach you have always claimed as your own but which was first described by me.
Finally, your joint letter of 5/4/98 does not refer at all to the memo dated 4/16/98 which deals with some points in the OGR letter and which was sent by email on that date to Mark Orazem. I would much appreciate a reply to this memo. In it, among other things, I requested that you provide corroborating details of your charge against me and Piterbarg of slander. Such details would consist of showing where any of the factual statements in our Note-added-in-proof were incorrect and malicious. As I mentioned in that memo, if any of our published facts were incorrect, we would certainly want to take remedial action. So far, you have not replied to any of the matters raised in the memo; I look forward to receiving a reply soon.
Sincerely,
(Signed by)
Ross Macdonald
******************************************
J. Ross Macdonald
354 Carolina Meadows, Villa
Chapel Hill, NC 27514
June 12, 1998
Professor Mark E. Orazem
Department of Chemical Engineering
University of Florida
Gainesville, FL 32611-6005
Professor Luis H. Garcia-Rubio
Department of Chemical Engineering
University of South Florida
Tampa, FL 33620
Dear Sirs:
Since I have had no reply from you to my memorandum (sent to Mark Orazem on 16 April 1998) or to my 5/26/98 response to your joint letter of 5/4/98 [referred to below as JL], I feel it desirable to write you once more about our on-going controversy.
In your joint letter [referred to below as OGR] of 2/19/98 to Roger Parsons, the editor of the Journal of Electroanalytical Chemistry, you cited the note-added-in-proof [NAIP], which appeared in the paper by myself and Piterbarg, JEAC 428(1997)1, and asserted that “his” (i.e, JRM’s) “comments” (in the NAIP) “do not properly describe the facts of the matter and therefore constitute slander.” I addressed this matter in the above-cited Memo and there requested that you substantiate this serious charge by showing that any of the facts I and Piterbarg included in the NAIP were incorrect. You have not responded to this request, although there has been more than sufficient time for you to do so. Since it would be wrong for you to make such a charge and not adequately support it, I am writing once more to discuss the matter and possible options which might allow us to conclude the affair.
In OGR, I have been able to identify only one place where you directly challenge a statement in the NAIP: that concerning my review of the original version of your paper (Durbha, Orazem, and Garcia-Rubio, J. Electrochem. Soc. 144(1997)48) [DOGR] where we said that your “MS be accepted only after extensive revision….” By contrast, in OGR you state that I advised in the review that the manuscript be rejected. As the original review shows, the NAIP statement is correct, and thus this is a clear misrepresentation on your part.
It should be obvious to you that a charge of slander which cannot (or is not) fully substantiated may itself be considered slander. But in addition to this slander against me and Piterbarg, in both OGR and JL you have accused me of several acts of “unethical behavior” and have thereby impugned my “personal and professional integrity” (quoted from JL). These are accusations which do not deal with the facts Piterbarg and I cited in the NAIP, but, if they are incorrect, also constitute overt slander against me personally.
I would like to point out to you once more that the NAIP was not intended to be either polemical and/or a “personal” “…attack on our (i.e., your) reputation.” (cited from OGR). We believe and can prove that it dealt with verifiable facts, ones you have failed so far to show are incorrect. In the NAIP, we did not accuse any of the authors of your paper of unethical behavior or of lack of scientific integrity, as you have characterized me. In OGR you state that in “his note-added-in proof, Professor Macdonald implies that we have plagiarized his work.” This is incorrect. First, it is wrong to ascribe the NAIP and its possible implications only to me, since it was a joint effort with Professor Piterbarg, who wrote part of it and approved all of it.
Second, please note that we never mentioned the word “plagiarism” in the NAIP. We neither accused you of it nor attacked any of you personally. While one interpretation of the facts we cited could imply the possibility of plagiarism, it is certainly possible that none of the three authors [DOGR] of the final published version of your paper benefitted from having the Macdonald-Piterbarg MS available during the three-month period during which you revised your MS. It is worth noting, in fact, that two of the three authors have, in OGR, stated that the M-P MS was not the source of the ideas used in the revision. In OGR it is also stated that, “We did not take the numerical technique from his paper. We have been using numerical integration … long before.” Unfortunately, this response obscures the point. That is not that DOGR learned about how to do numerical integration from our MS, a nonsensical suggestion, but, instead, that our MS, completely unlike the original version of the DOGR MS, first introduced the idea of using numerical integration results analytically to produce a valid proof.
Let me now summarize the personal attacks you have made on my scientific integrity and your accusations in OGR of unethical behavior on my part. Note again that these matters do not deal with the facts we cited in the NAIP so are unrelated to your original charge of slander by me and Piterbarg but constitute additional slanders of me if they are incorrect.
I have already pointed out here and in the Memo the incorrectness of your charge that as a reviewer I recommended rejection of the original version of your MS. Further, your accusation in OGR that my “actions constitute a serious violation of the integrity of the review process” has been addressed in the Memo and shown to be without merit.
Next, your accusation in OGR (repeated in JL) that in the Boukamp-Macdonald paper (Solid State Ionics 74(1994)85), (erroneously also referred to in OGR as a 1993 paper), as well as in the Macdonald-Piterbarg paper cited above, we “claimed credit for our (i.e., you and your co-workers’) idea of using distributed time constant models to assess consistency with the Kramers-Kronig relations…” has been addressed both in the Memo, and more completely in my letter to you both of 5/26/98. We did not claim such credit since we cited some of your relevant prior work in the field. I regret, as I have said, that you feel that we did not give you sufficient credit for this idea in the Boukamp-Macdonald paper (the Macdonald-Piterbarg paper did not deal with this method in any event), but it turns out, as discussed in the 5/26/98 letter, that you and your co-workers have been remiss for many years in never giving me credit for the original1987 idea!
Finally, I would like to suggest that it would be to everyone’s benefit if we could conclude this affair quickly and privately. To do so, I request that you write me and include:
(a) Either a full justification of your original charge of slander or a statement that you withdraw it.
- A statement in which you withdraw your accusation that I and my co-authors claimed credit in our 1994 and 1997 papers for originating your idea of the measurement model- KK assessment approach. As we now see, it would have been appropriate for us to cite my own earlier work, but that was not done.
Your withdrawal of the charge that I violated the integrity of the review process.
If I receive such a letter, I will consider the matter closed. If, on the other hand, I do not hear from you by 17 July 1998, I will have to consider other options – such as going public about the whole affair and your failure to respond properly to the matters you raised in OGR and JL. I hope that will be unnecessary, and I look forward to receiving your response soon.
Sincerely,
(Signed by)
Ross Macdonald
***********************************
No answers to the above two letter have ever been received.