Talk:Koch's postulates

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Older comments[edit]


I am currently away from my reference books at my home and will fix the above text as soon as I am able.



The postulates are wrong. All say the exact opposite of what Koch proposed. Comments written below that complain of confusion have merit. Please correct this. —Preceding unsigned comment added by (talk) 15:15, 10 May 2010 (UTC)

I don't really understand this. Does the second postulate say that the pathogen must be able to be grown in a a pure culture, or this this merely a step before postulate 3? Also, my biology textbook thinks that postulate 4 continues with the two cultures being shown to be identical. Any advice from knowledgeable persons is much appreciated! Thanks. -- postglock 16:58, 26 August 2005 (UTC)

There is some information here regarding the example of leprosy not being able to be grown in a pure culture. I am still unsure of the implications though. I suppose it suggests Koch's postulates are just that, postulates with exemptions, not laws? -- postglock 17:07, 26 August 2005 (UTC)

they are postulates, not laws. I have updated the article with a little more information, as there are always exceptions to rules... --Grcampbell 19:43, 26 August 2005 (UTC)

There are multiple variants of the postulates on the Internet, some more coherent than others. Which is the original? Here is one sample:

  1. The organism must be found in all animals suffering from the disease but not in healthy animals.
  2. The organism must be isolated from a diseased animal and grown in pure culture.
  3. The cultured organism should cause disease when introduced into a healthy animal.
  4. The organism must be isolated again from the animal into which it was introduced.

Could someone who has Koch's 1890 paper please check this for us? --Una Smith 15:07, 5 July 2007 (UTC)

I've got access to the paper, but it's not translated and Ich spreche Deutsch nicht so gut. MastCell Talk 17:45, 5 July 2007 (UTC)
MastCell, without translating, does the paper match the German wikipedia version of Koch's (1890) postulates? How about we get a translation of their "modern" postulates? --Una Smith 20:46, 5 July 2007 (UTC)
In Inventing the Aids Virus (1996), German author Peter Duesberg writes that in 1884 Koch published a paper on tuberculosis that “spelled out the key criteria for proving a microbe guilty of causing a disease." Duesberg states them in English as follows:
"First, the germ must be found growing abundantly [my italics] in every patient and every diseased tissue.
Second, the germ must be isolated and grown in the laboratory.
Third, the purified germ must cause the disease again in another host.”
If asymptomatic carriers don't have the germ in abundance then the postulates hold for them too. Although I'm not a scientist, I suspect this is indeed the case.
Regarding cholera, Duesberg writes that Koch isolated the correct bacterium, but couldn't induce the disease in test animals, and so used statistical correlations as proof instead of laboratory results. Duesberg theorizes that Koch used animals that had been vaccinated by natural infection. Presumably natural infection results in vaccination in many animals and people, thus making them asymptomatic carriers. Scientists have since induced cholera in unimmunized animals, Duesberg writes. 21:44, 30 November 2007 (UTC)
Peter Duesberg's view of the postulates should be interpreted cautiously, given his agenda, particularly as his application of them is well outside the medical mainstream. MastCell Talk 21:57, 30 November 2007 (UTC)
I would rephrase that by replacing cautiously with strictly. 22:25, 30 November 2007 (UTC)
Actually, I'll rephrase it thus: Duesberg's views should be given the same amount of weight on Wikipedia as they are given by the expert scientific community. Which is to say none. MastCell Talk 22:32, 30 November 2007 (UTC)
I am very confused in trying to reference these postulates. The article maintains that they are formulated in 1890, but there is no 1890 paper in the reference section. Further, the first citation for the 1876 paper seems to contain the postulates in the section concerning methods and techniques. I do not speak German well enough to confirm this. Is the 1890 date incorrect, and where is the 1890 paper? — Preceding unsigned comment added by (talk) 20:55, 5 February 2014 (UTC)

It appears this section is being adjusted to oppose Duesberg rather than to provide information on Koch's postulates.Nukeh (talk) 07:13, 16 February 2008 (UTC)

It's been "adjusted", in keeping with Wikipedia's policies, to avoid giving undue weight to a fringe view which is already covered in great detail elsewhere on Wikipedia. MastCell Talk 21:59, 16 February 2008 (UTC)
Can you explain to me why an HPC plus HIV contaminated needle stick is treated as an HPC stick? Duesberg's "weight", as you cite above is not "none". Weight = 0, is a mathematical / physics concept. If you mean Duesberg is wrong, you might be right. But that is not for us to decide as editors. I still assert that a discussion of Duesberg and HIV=AIDS has no place behind an article on Koch's postulates. This indicates a biased agenga to fix-up Koch for referencing from other articles in WP.Nukeh (talk) 19:12, 18 February 2008 (UTC)
I'm entirely unclear on what you're arguing or suggesting. Duesberg is not mentioned in this article. Are you agreeing or disagreeing with that state of affairs? MastCell Talk 20:11, 26 February 2008 (UTC)

An authoritative English translation of the relevant portion of Koch's 1884 paper is available in "Milestones in Microbiology: 1546 to 1940", translated and edited by Thomas D. Brock. American Society for Microbiology Press, 1999ISBN13: 978-1-55581-142-6 Accessible on Google Books. —Preceding unsigned comment added by Mdalbey (talkcontribs) 03:39, 21 March 2010 (UTC)


I've changed the text (see history) to say he abandoned the second part of the fiorst postulate. I.E. the organism must still be in all affected animals, but not neccesari;y absent in healthy one. Rich Farmbrough 23:18 3 March 2006 (UTC).


This is a stray, I don't know where it belongs. --Una Smith 21:06, 5 July 2007 (UTC)

  • Koch R. Über die Ätiologie der Tuberkulose. In: "Verhandlungen des Kongresses für Innere Medizin. Erster Kongress, Wiesbaden 1882".


This article appears to have been adjusted such that anyone investigating HIV=AIDS theories would find nothing here for support. Nukeh (talk) 07:09, 16 February 2008 (UTC)

Then it would be accurate. MastCell Talk 21:58, 16 February 2008 (UTC)

Do you mean that Koch's postulates (today) are the same as Koch's postulates (when he wrote) from the proper historical perspective? Or do you mean that anyone who believes there is the very slightest question as to whether HIV = AIDS would find nothing here? If the latter is what you mean to imply, then why do Duesberg's views get mentioned on this article's discussion pages? If Koch's postulates change over time, then I suggest we disambiguate into the historic Koch's postulates (this article) and place modern modifications of Koch elsewhere.

Please try to follow this analogy: If one changes a single note in a Mozart piece, it is no longer Mozart. As one changes a large number of his notes in a musical piece, the work becomes substantialy different to the point that it might be considered a new work. The same would hold true in mathematics, where, in fact, changing a single 'note' is probably an error. Biology is not as solid as mathematics except in the historical perspective, where historians (editors like us) do not change the original work, whether it is right or wrong. —Preceding unsigned comment added by Nukeh (talkcontribs) 18:30, 18 February 2008 (UTC) Nukeh (talk) 18:57, 18 February 2008 (UTC) neat bot! Nukeh (talk) 18:57, 18 February 2008 (UTC)

This talk page is intended for discussion of concrete and specific improvements to this article. I'm not sure what you're trying to suggest in that regard. AIDS denialism is a fringe view which is already given more than its due weight on Wikipedia in the articles dedicated to it; it really has no place here. MastCell Talk 22:54, 18 February 2008 (UTC)
The name "Duesberg" is on this discussion page more frequently than "Koch", and I agree that Duesberg is a fringe view, but he has more than his due weight here.Nukeh (talk) 07:41, 26 February 2008 (UTC)
That this talk page is a magnet for AIDS denialism is unfortunate. However, WP:WEIGHT is determined by proportionate acceptance of a view by experts in the relevant field, not by the number of mentions of a concept on a Wikipedia talk page (God forbid!) MastCell Talk 20:09, 26 February 2008 (UTC)

Doubts about authors[edit]

The article says the authors are Robert Koch and Friedrich Loeffler. I read somewhere Friedrich Henle. Wich is correct? Henle was Koch student. Khullah (talk) 02:46, 1 September 2009 (UTC)

List of Diseases[edit]

This talk page is for discussing specific reliable sources and concrete improvements to the article. It is not a forum to expound one's personal opinions on various infectious agents. Discussion can be reopened with reference to actual reliable sources and concrete changes to the article.
The following discussion has been closed. Please do not modify it.

I think it would be interesting for readers to have a list of the most well-known diseases that either pass and fail the postulates. Does anyone have any objection to me starting a section for each? BruceSwanson (talk) 16:39, 20 July 2010 (UTC)

I'm not sure how useful that would be, since the postulates are not ironclad and don't take into account modern knowledge. "Failing" the postulates does not rule out causation, and they've modified post hoc several times (including by Koch himself) to address cases where an organism clearly caused disease but still "failed" the postulates (e.g. tuberculosis, cholera, typhoid). In any case, perhaps you could bring the sources you intend to use to create this list here for discussion first? MastCell Talk 16:59, 20 July 2010 (UTC)
The postulates are really just a rule of thumb - actual causation is determined by a full and proper statistical analysis or whatever. My guess is that the issue is not a particularly significant part of the topic, but I could be convinced otherwise. Did you have any particular sources in mind for this section? - 2/0 (cont.) 21:09, 21 July 2010 (UTC)

But you aren't arguing that some non-infectious diseases follow Koch, right? You accept that scurvy and pellagra never follow Koch because they cannot do so. If you're with me so far, then it is clear that the postulates are absolutely unequivocal under certain conditions (described by the postulates themselves), which is to say, for certain diseases. Right? So now the problem is to extend the list of well-known (and hence interesting) diseases for which Koch's postulates are final. But the list of diseases that never follow Koch would always be diseases since proved (using a combination of the postulates, new technology, and statistical analysis) to be non-infectious. Such diseases, then, always follow Koch by never satisfying the postulates. Likewise, diseases accepted to be infectious could only contradict the postulates for known and very conditional reasons -- natural inoculation (in the case of Koch's anthrax experiments), or the nonabundant amount of bacteria or virus in a patient's blood or tissue (like most unvaccinated people infected with polio). Koch stressed that the infectious agent must always be of sufficient number -- and so they always are, where symptoms are present. Right? Well, except, of course, for HIV and Hep C viruses -- and prions. They all fail Koch's postulates by [apparently] causing disease in the absence of abundant infectious agents. That's three. Can you think of any others? BruceSwanson (talk) 02:00, 27 July 2010 (UTC)

While I'm impressed at the progression from AIDS denialism to hepatitis C and prion denialism, could we avoid discussing our personal interpretations of Koch's postulates (per the talk page guidelines) and stick to what reliable sources have to say? As I'm sure you know, HIV does fulfill Koch's postulates, at least in the opinion of scientists who actually do research on the subject. What reliable sources do you have in mind to substantiate this discussion? If the answer is "none", then let's archive it now and save our breath. MastCell Talk 04:31, 27 July 2010 (UTC)
Bruce's mind must be fertile indeed, since ideas take root so strongly that even a blizzard of facts cannot uproot them. Tim Vickers (talk) 18:12, 27 July 2010 (UTC)

Koch describes a process of rigorous clinical-testing only. So HIV fails Koch by definition, specifically #3: The cultured microorganism should cause disease when introduced into a healthy organism. HIV doesn't induce AIDS in animals (we know that much), and law and ethics forbid using human candidates. AIDS orthodoxists claim that numerous cases of accidental exposures (needle jabs, etc.) have been followed by AIDS (or at least an HIV+ test-result). On that basis alone can HIV conform to Koch. However, the postulates aren't about anecdotal data, which is also a source of counterclaims that many lab and hospital personnel have been needle-infected by HIV and never come down with AIDS (or a positive HIV-test) at all. Logic excludes both from consideration in testing the postulates. (It should be noted that medical personnel who have been jabbed by an HIV-contaminated needle are now urged to immediately begin an antiretroviral regimen, including the chemotherapeutic agent AZT.) Finally, not enough is known about Idiopathic CD4+ lymphocytopenia to exclude it as a possible Koch disqualification of HIV.

You wrote above that tuberculosis, cholera, and typhoid fail the postulates while clearly being infectious. Again, Koch himself insisted on a sufficient quantity of bacteria -- one of the modifications you referred to above. A healthy human body plays host to an abundance of potentially lethal infectious agents -- including tuberculosis. A healthy immune-system keeps them all in check, even in the face of other serious illnesses and injury. Does that mean that all those pathogens fail Koch as well? Are patients ill with these diseases ever found with anything but levels of infection sufficient to explain the illness?

I think the simplest approach is to list only those diseases, bacterial and viral, that clinically fail Koch's Postulates. So I nominate HIV, Hep C, and prion diseases.

2/0, any comments? BruceSwanson (talk) 18:20, 27 July 2010 (UTC)

HIV is known to satisfy Koch's postulates, as documented in the scientific literature (e.g. PMID 8902385) and by expert scientific bodies (e.g. [1]). I've asked you to provide sources for your claim to the contrary, but you've responded only with your personal viewpoint (marked by both logical and factual errors). You're misusing this talk page, and it's time to break the cycle. If you don't make some sort of effort to produce reliable sources with some relevance to the article content, I'm going to archive this thread in the next 24 hours or so. MastCell Talk 18:37, 27 July 2010 (UTC)
Indeed. You have been asked specifically several times to provide your sources, but have not yet done so. Articles such as [2] specifically refute your ill-informed suppositions, so unless you can finally cite some authoritative sources, I agree with MastCell that there is nothing more to say here. Tim Vickers (talk) 18:43, 27 July 2010 (UTC)

The sources MastCell and TimVickers linked to (above) aren't satisfactory: Vickers' link duplicates MastCell's (PMID 8902385), both to a mere title, not even an abstract; but the second link is to a reliable secondary-source. However, it claims that HIV conforms to Koch by citing conclusions drawn from events not part of a rigorously controlled, clinical setting: accidental needle jabs, and the David Acer / Kimberly Bergalis case. As I pointed out earlier, Koch's postulates are based solely on clinical findings. They are exercises in pure laboratory science and conclusions drawn logically from them. BTW, that same source also contains the following remarkable statement: Koch's postulates also have been fulfilled in animal models of human AIDS. Chimpanzees experimentally infected with HIV have developed severe immunosuppression and AIDS. Yet the primary source provided ("O'Neil" et al), states Chimpanzees are susceptible to infection with human immunodeficiency virus (HIV)–1; however, infected animals usually maintain normal numbers of CD4+ T lymphocytes and do not develop immunodeficiency. Not exactly a strong correlation to Koch.

It should be understood the the postulates are what they are and nothing else. If a disease doesn't conform to them, it doesn't conform. Period. That doesn't necessarily mean it isn't infectious. It doesn't mean it won't be found later to conform. It doesn't mean it wouldn't pass a modified version of them. It simply means it failed the logical precepts historically known as "Koch's Postulates". That's it. BruceSwanson (talk) 00:38, 28 July 2010 (UTC)

What are Koch's Postulates?[edit]

General questions and discussion about this topic, please use the Wikipedia:Reference desk instead, per WP:Talk.
The following discussion has been closed. Please do not modify it.

Here is a link to a reliable source that claims HIV satisfies Koch's Postulates. Look at the section entitled Evidence That HIV Causes AIDS. You can see that in justifying HIV's satisfaction of postulate #3 (Transfer of the suspected pathogen to an uninfected host, man or animal, produces the disease in that host) it describes data obtained from uncontrolled sources -- accidental needlepricks, the David Acer/Kimberly Bergalis case, etc. (Note as well that the classic four postulates are condensed to three.)

Would it be original synthesis to state that data from such uncontrolled sources from outside the controlled laboratory-environment cannot logically satisfy the postulates, if a reliable source includes such data in its claim to have done so?

Which takes precedence: the logic of the postulates themselves as derived form their literal words, or what a reliable source says about them? If it's the latter, shouldn't this article state that data derived from outside sources has successfully been used to justify conformity to the postulates, and then include the linked reference (above) as a footnote?

I'm particularly interested in the opinions of real-name editors who work in the scientific field. BruceSwanson (talk) 20:06, 28 July 2010 (UTC)

The postulates do not require that evidence come from a "controlled" source, whatever novel meaning you've assigned to the term. They only require (well, suggest) that an organism should be capable of causing disease if introduced into a healthy organism. HIV isolates are clearly capable of that, so the NIH statement that you linked seems quite sound. I found the remainder of your post difficult to parse, but it doesn't seem that it contains any specific changes to the article text that would be justified. Maybe you'll get a different answer from my "real-name" comrades, though. MastCell Talk 20:53, 28 July 2010 (UTC)
In Wikipedia we summarize what the source says and do not interpret it in novel ways ourselves. Since this source says that the postulates have been fulfilled, and the source is reliable, we can say that the postulates have been fulfilled and cite this source to support the statement. Since you have still not provided any sources to justify your personal convictions about Koch's postulates, you are continuing to abuse this talkpage. Please remember that Wikipedia is not a forum. Tim Vickers (talk) 20:57, 28 July 2010 (UTC)
Echoing Tim and MastCell. BS is violating WP:OR by attempting to argue for a claim that is clearly contradictory to the scientific consensus. This leads to WP:NPOV which says we give weight to the scholarly majority and can safely ignore fringe pseudoscience like HIV denialism a priori. Finally, BS' use of the talk page to argue for an inappropriate application of sources to further an AIDS denialist agenda is a WP:FORUM and WP:TP problem. So we're yet again, done. Bruce, HIV causes AIDS. That's the scientific consensus. If you don't like that wikipedia adopts that position, you can attempt to convince the community to change its stance on pseudoscicence, the scientific community and original research. Until that happens, please don't bother pushing a dishonest and discredited agenda. You're not casually interested in Koch's postulates, you're interested in misapplying them to prove HIV does not cause AIDS. We're not interested. WLU (t) (c) Wikipedia's rules:simple/complex 18:14, 29 July 2010 (UTC)

All this brings to mind We_are_Seven. What are Koch's Postulates? I say They are four. Tim Vickers, where in this article could the reliable source under discussion be used? Shall it reference a statement that Koch's Postulates are sometimes condensed to three? But then, are they "Koch's Postulates" if they are only three? Aren't "Koch's Postulates" four by definition? Therefore doesn't one have to satisfy all four if you are to claim success in conforming to that logical entity known as "Koch's Postulates"? I'm interested in your opinion on this since you are a professional in the field.

I'm unable to take seriously your charge that my questions constitute abuse of the talk pages. BruceSwanson (talk) 18:13, 31 July 2010 (UTC)

This article lists only 3 postulates - the 3 that the NIAID website describes in reference to HIV. As an aside: "Even at the time they were presented, the Henle-Koch postulates were never recommended as rigid criteria of causation and failed to apply to many diseases at the time when a causal relationship seemed almost unequivocal; Koch himself felt that fulfillment of the first two postulates might be sufficient evidence of causality." One wonders at the effort to turn these into rigid criteria of causation in 2010 when they were not considered such by their original authors in 1840. MastCell Talk 20:21, 31 July 2010 (UTC)

MastCell wrote, above: "This article lists only 3 postulates - the 3 that the NIAID website describes in reference to HIV." Would anyone willing to replace the current four postulates in the article with either set of three, and cite the respective source as a reference?

The NIAID source is here. I'll leave it to hypothetical readers to compare both three-postulate versions and decide whether they are quite the same. The NIAID version:

1. Epidemiological association: the suspected cause must be strongly associated with the disease. 2. Isolation: the suspected pathogen can be isolated - and propagated - outside the host. 3. Transmission pathogenesis: transfer of the suspected pathogen to an uninfected host, man or animal, produces the disease in that host.

You can see why MastCell wrote, above, of the postulates in general: They only require (well, suggest) that an organism should be capable of causing disease if introduced into a healthy organism. The postulates not as self-evidently logical and clear, but rather as a kind of clinical affirmative-action program.

Koch/Henle's 1840 version, from the PDF:

1. The parasite occurs in every case of the disease in question and under circumstances which can account for the pathological changes and clinical course of the disease. 2. It occurs in no other disease as a fortuitous and nonpathogenic parasite. 3. After being fully isolated from the body and repeatedly grown in pure culture, it can induce the disease anew.

You can see why Koch thought that the first two might be sufficient. But can anyone imagine that reports from the field would be sufficient to satisfy them? This is to ask a more fundamental question that deserves further discussion here:

If the postulates are to be considered merely advisory, and not rigid criteria of causation of infectious disease, then one wonders why the NIAID source in question claims that HIV has fulfilled them. Why bother? After all, there are perfectly respectable infectious diseases, viral and bacterial, for which, in the words of the PDF article's discussion of EBV, "current data are still inadequate to fulfill the Henle-Koch postulates". In the particular case of HIV, we should be aware that the NIAID source claims #3 is satisfied on the basis of reports of lab-workers accidentally contaminated; and the Acer/Bergalis case. What isn't mentioned is whether the workers later began a regimen of AZT, as Bergalis did. Did some take the treatment? Did all? Did any? Did any die or did they recover? And what exactly were their symptoms -- respiratory, neurological, immunological? Can anyone seriously argue that such questions needn't be answered before determining whether their cases can be used as clinical data to claim conformity with the postulates? We can be sure of one thing: that none of the alleged victims were enrolled in a study denying them antiretrovirals in order to see whether they in due course died of AIDS. BruceSwanson (talk) 18:22, 2 August 2010 (UTC)

You know, in the late 1800s, Alfred Russell Wallace got into a dispute with a member of the Flat Earth Society. The Flat Earther demanded conclusive proof that the Earth was round, and wagered £500 that Wallace couldn't provide it. Against his better judgment, Wallace took up the challenge - after all, the Earth is round, and it isn't that hard to prove, and £500 was a lot of money in those days. Wallace took the Flat Earther to the Old Bedford River, the Flat Earther's home territory, and was able to demonstrate conclusively the curvature of the river's surface.

What happened next is instructive. The Flat Earther refused to honor the wager, raising a number of post hoc objections which amounted to goalpost-moving. Having lost the battle in the reputable arena, the Flat Earther developed a personal animus toward Wallace, and engaged in a years-long campaign of defaming him and ultimately making death threats against him. Wallace prevailed in court, but the ultimate cost to him was clearly disproportionate. Wallace's engagement with the Flat Earther was a failure on a more fundamental level - it failed to convince any of the Flat Earthers of the error of their worldview - because the Flat Earthers were not capable of being convinced by rational, objective, factual discussion or demonstration. It was easier across the board for the Flat Earthers to embroider their conspiracy theories and increasingly bizarre objections than to rationally evaluate Wallace's demonstration and recognize that they were fundamentally wrong.

In Texas, and perhaps other parts of the South, Wallace's experience would be summed up thus: "Never wrestle with a pig, because both of you will end up dirty, and the pig will enjoy it." I apologize for the tangent. To return to the point, this talk page does not exist as a forum for you to propound your personal philosophy on HIV/AIDS. Your userpage indicates that you have a blog, so if you'd like to publish your personal exposition of Koch's postulates, that might be the best venue. Here, let's stick to sources. If you are aware of any which address (or even raise) the questions you mention, please present them. Otherwise, I'll probably collapse this discussion, like the last one, as another abuse of this article talk page. MastCell Talk 18:45, 2 August 2010 (UTC)

This is a reliable source. The following is a question about it:

BruceSwanson (talk) 15:24, 3 August 2010 (UTC)

HIV fulfillment of the postulates[edit]

My last edit was to show that the footnoted source finds three postulates sufficient to prove causation. Thus it is in agreement with its sentence: ... fulfillment of all four postulates is no longer required to demonstrate causality. Operative67 (talk) 00:09, 31 July 2010 (UTC)

Weird phrasing, "in support of HIV/AIDS denialism" / BTW, amendments to Koch's postulates?[edit]

The assertion that HIV causes AIDS does not follow Koch's postulates[13] in support of HIV/AIDS denialism. It's probably some edit left-over of something that probably intended to mean something like "HIV/AIDS denialists assert that the association of HIV and AIDS doesn't fulfill Koch's postulates, even though the scientific criteria for such associations have been expanded since then, so on and so forth". By the way, I recall having read something, I guess from non-HIV/AIDS denier sources, along the lines of an "updated version" of such postulates, with a few additions, actually motivated by HIV/AIDS research. — Preceding unsigned comment added by (talk) 03:11, 19 April 2016 (UTC)