“Is a tomato a fruit or a vegetable?” says the teacher. “A vegetable!” shout the children. “Wrong,” replies the teacher with a smug look, “It’s…It is a fruit”. Both are wrong, although the teacher bears more responsibility for propagating false fruits and for using rote humiliation to substitute for learning.
The right answer is that a tomato is both a fruit and a vegetable as well as only a fruit and only a vegetable. The choice of question, use of words with multiple definitions without qualification and the purported correct answer says a lot about how rigid thought patterns and our use of language manipulate our beliefs about science.
The most specific meaning of ‘fruit’ is the botanical: “The seed of a plant or tree, regarded as the means of reproduction, together with its envelope”. By this definition, the tomato is clearly a botanical fruit, although the little piece of stem and sepals that are often attached must be removed for purity of this definition to be maintained. If we want to get picky we could note that people who remove the skin, seeds or central core of the tomato, apart from being wasteful, are also no longer dealing with a (botanical) fruit.
This definition of fruit is fundamental to a hierarchical , anthropocentric view of botany. Enclosed seeds are considered a mark of the higher plants (angiosperms), just as gestation, growth of the fetus within its maternal envelope, is a mark of superiority for mammals. This distinguishes them from all the ‘lower’ plants or animals that do not fully enclose their offspring.
The most general definition of the word ‘fruit’ includes virtually everything that is edible. The ‘fruits of the earth’ would include animals, but would exclude many inedible fruits (defined botanically). By this definition a tomato is a fruit, but so are cows and carrots. More restrictive definitions include only plant materials that are edible.
The children in the classroom are most likely unconsciously using yet another flavour, the gastronomical definition, where a fruit is a food material that can be used for dessert, as compared with a vegetable defined as a plant material that is used for the savory courses of a meal.
By the botanical definition a tomato is a fruit. If botanical definition is used, then the question, “Is a tomato a botanical fruit or a botanical vegetable?” makes no sense as there is no botanical definition of a vegetable.
There is nothing wrong with a gastronomical definition. Humans could get along quite nicely without plant taxonomists and anatomists (at least in the short run), but where would we be without cooks? Classifying cooking materials by their usage in the kitchen is useful as long as nobody pretends that this classification sheds any light on the evolution of the species, any more than it would make sense to assume that nux vomica, being a botanical fruit, is edible.
By the gastronomical definition, the tomato is clearly a vegetable as it is never, well rarely (I was once exposed to a chocolate cake using tomato soup as a base), used in the dessert, sweet or fruit course of a meal. Even the US Supreme Court agrees. During an 1880s trade dispute its justices ruled that the gastronomic definition of vegetable applied to the tomato more than the botanical definition. Consequently, duties on the importation of fruits did not apply to tomatoes.
A fairer question would be to ask “Is a tomato a botanical fruit or a gastronomical vegetable?” Apart from giving the game away it would invite the answer “Both”. This is abhorrent to the atomistic urge to classify everything neatly into one compartment or another. Somehow we believe that it is scientific to divide the world into mutually exclusive boxes and unscientific to claim an area of greyness between two extremes.
Classification can be very useful when it reflects an underlying reality. In this case the best classification of a tomato depends on whether one is trying to highlight the anatomical similarities of the angiosperms or the useful of plants in the kitchen. When the harried cook asks “What do we have in the way of fruit for supper?” everyone but the most annoying pedantic child would know that the right answer might be “Rhubarb and Pineapple” and certainly not “Hazelnuts, Green Pepper and Black Pepper”!
When studying the relationships between plants it is very important to understand the botanical definition. The anatomy of a fruit usually gives important clues to the family that a plant belongs to, allowing a botanist to partially classify a newly encountered species upon first sight. Although we can never know precisely how species evolved, a classification based extensively on the complex reproductive portions of plants, is our best guess.
It is interesting that the tomato is almost always used for this question, rarely other gastronomical vegetables that are botanical fruits. A google search, for example, turned up 295 pages with the exact words “Is a tomato a fruit or a vegetable”, 11 for “Is a cucumber a fruit or a vegetable” and none for “Is a * pepper a fruit or a vegetable”, where * is ‘green’, ‘red’, ‘sweet’ or nothing.
If I go through the vegetable drawer in my refrigerator, there are many examples of vegetables that are also fruits. Sweet and hot peppers are clearly fruits as are the many representatives of the Cucurbitaceae such as zucchini, squash as well as cucumber. Eggplants, corn, peas, beans and okra are clearly also fruits. Lemons and limes usually reside in this drawer (in my house at least) although even in the kitchen they straddle the definitional fence, being used for savory dishes, especially in Mediterranean and Middle Eastern cooking, as well as for desserts.
The reason for the use of the tomato instead of one of many other vegetable fruits is because the question is part of a legend, a modern version of a fable handed down from one generation to another. If teachers were really trying to make a point about scientific definitions they would go through the exercise of classifying the contents of a typical vegetable drawer and make sure their students knew what the distinguishing characters of a (botanical) fruit were. Having students know only that a tomato is a fruit is a burden of trivia. Teaching students to botanically classify vegetables would be to achieve real learning.
While the average vegetable drawer is chock full of fruits, the situation appears simpler on the other side of the fridge. Oranges, grapefruits, apples, pears, bananas (well, okay, they go black if you keep them in the fridge, but certainly this is the drawer for them if that was not the case), peaches, apricots and grapes are (less the attached pieces of stem and flower) clearly botanical fruits as well as gastronomical fruits. But it only takes one example to make rhubarb of the notion of purity in the fruit drawer. The pineapple is, botanically speaking, not a fruit because the edible portion includes the entire flower (petals and sepals) not just the seeds and its covering. The lowly rhubarb is, of course, a piece of petiole (leaf stem), not a fruit, even though it is a common ingredient in fruit pies.
If the rest of the kitchen is examined, the situation gets worse for botanical purists. Clearly nuts, in their shell, are fruits. The edible part is usually the seed, the hard shell of the sunflower seed, pistachio, pecan, walnut or brazil nut is the enclosing envelope that defines it as the fruit of a higher plant. Except, of course, for pine nuts, which are not botanical nuts because they are a product of one of those lower plants, the gymnosperms.
The spice drawer is also full of fruits, or at least parts of fruits. Black pepper, dill seed [sic], coriander, cumin, fennel, mustard, caraway, vanilla and juniper berries (oops, there are those darned gymnosperms again).
Perhaps what teachers who repeat this legend are really trying to impress on their children is that scientific terms are more precise than those from everyday life. Common names of plants and animals are notoriously unreliable. Robins in England, Canada, Australia and India are very different birds. All they have in common is that they exist in an area that was colonized by the British who pined for the symbols of home, and likely named the first red (or close to red) breasted bird they saw as a robin, without any consideration for taxonomic principles.
Scientific terms can often be very precisely defined. Physical constants that are known to a large number of significant digits leap to mind at the speed of light (measured in a vacuum, of course). The danger is, that by implying that a term is scientific, people, including scientists, will leap to the assumption that the term is precisely defined. Not every term proposed by scientists is scientifically defensible.
The terms HIV and AIDS come to mind (you knew that I was going to get here eventually). Just about everyone knows that HIV is a virus, that AIDS is a disease and that HIV is the sole cause of AIDS. Very simple, very precise. But, very probably wrong.
AIDS, as a term, is the easiest to throw rotten tomatoes at. Apart from the simple, and not very useful, definition of it being the disease caused by HIV, AIDS is really defined by organizations such as the CDC and the WHO. Their definitions are shockingly different.
The CDC definition of AIDS has changed several times since the term was first coined in the early 1980s. Back in the days of GRID (Gay-Related Immune Deficiency) the definition included only Kaposi’s sarcoma (a skin cancer) and Pneumocystis carinii Pneumonia (PCP; a serious lung infection). Previously found only in elderly or immune suppressed people, these diseases were suddenly being found in young, previously healthy men, almost all homosexual.
Gradually it was found, in this highly promiscuous subset of gay men, that other diseases were also being found, and these were lumped into the definition until, by 1987 the definition of AIDS included (even in the absence of a positive HIV test) certain types of candidiasis, cryptococcosis, cryptosporidiosis, cytomegalovirus disease, herpes simplex, lymphoma, lymphoid interstitial pneumonia (in a child under 13), Mycobacterium avium or kansasii, progressive multifocal leukoencephalopathy, toxoplasmosis of the brain (in an infant) as well as the old standbys, Kaposi’s sarcoma and PCP. With a positive HIV test in hand, the definition was even broader, including tuberculosis, some types of non-Hodgkinn’s lymphoma and ‘wasting’ (e.g. loss of more than 10% of body weight and chronic diarrhea).
By 1993 , the definition had been further broadened to include a non-disease as part of the definition of AIDS. Low CD4 (Helper T cell) counts or a skewed CD4/CD8 ratio along with a positive HIV test without any AIDS-defining illness could result in a definition of AIDS. By 1994 just over half the new AIDS diagnoses were in this category, and by 1997 about 65% of AIDS diagnoses were in people who, by definition, had no AIDS-defining illness and who may have been perfectly healthy. This might have started to become embarrassing for the CDC so, in the surveillance reports for 1998 and later, this statistic disappeared. When I, at the urging of Ukrainian statistician Vladimir Koliadin, wrote to the CDC and asked for the raw data that would allow us to obtain this information (and many other interesting statistics) we were told that the CDC, monster bureaucracy that it is, simply did not have the time or money to provide us with the information.
It is clear then that, in the US at least, AIDS is no longer a disease, it is what it is, nothing more, nothing less. The general population fervently believes that it is a disease, as do most medical workers, but the published definition and related statistics clearly show that it is not. It is certainly not a single disease, not even a cluster of related diseases, and not necessarily associated with any illness (although always with the prediction of illness and death in the near future of its victim).
In the Third World, the WHO definition of AIDS usually applies. For this definition, illness is a requirement, but not HIV. AIDS is diagnosed if at least two ‘Major signs’ and one ‘Minor sign’ are present. Major signs are:
The list of minor signs is longer, but includes cough persisting for more than a month (except in people with recognized tuberculosis) and itchy dermatitis. Kaposi’s sarcoma and cryptococcal meningitis are trump cards, resulting in an AIDS diagnosis without the presence of major or minor signs. The only exception to this rule is that if there is a reason for these signs that is known and is unrelated to HIV, then the illness is not classified as AIDS.
The WHO admits that “HIV-negative tuberculosis patients could be counted as AIDS cases because of their similarity in clinical presentation.”
So, in Africa, the definition of AIDS is as different from AIDS in America, as 1993 American AIDS differs from 1984 American AIDS. It can be diagnosed with illness without HIV, whereas AIDS in America now is most commonly diagnosed with HIV, but without illness.
Without HIV being visible it is obviously impossible to determine its presence directly. There is nothing wrong with detecting something that is too small to be visible indirectly, but there has to be a tight chain of logic between the signs and the actual thing.
In the case of HIV the sign of its presence is most commonly an antibody test. This is strange because antibodies are not a reliable indicator for the presence of a pathogen. An antibody used to be taken as a sign of pathogen past, not necessarily pathogen present or future. Antibodies were taken as a good thing, a sign of immunity. Childhood diseases would be suffered once, followed by lifelong immunity. Vaccination is based on the stimulation of this lifelong immunity without having to experience the disease even once.
But with HIV (and some other modern diseases) the presence of an antibody is seen as proof of the presence of active virus. By accepting the hypothesis that HIV cannot be overcome by the natural human immune system, the presence of antibodies must mean that HIV is still present, armed and dangerous. This has a very profitable side effect, as it also means that the cure cannot come from within, but only from without, via expensive (and highly toxic) antiretroviral drugs.
There are other methods for detecting HIV, such as the Polymerase Chain Reaction (PCR or ‘viral load’), culturing and antigen tests. The major barriers to their use are that they are more expensive and time consuming and (with the exception of PCR) would classify far fewer people as HIV infected..
Antigen tests (usually p24, believed to be one of the structural proteins of HIV) are positive in only about half of the people who are HIV antibody positive , , , , , . HIV antigens have also been identified in the placentae of HIV-negative women . The use of this test is now discouraged except in screening blood and plasma .
The explanation for this discrepancy is that the HIV antigens are bound in immune complexes, although that does lead to the question of why this is not accepted as immunity. For various reasons, antigen tests have been displaced, since the early 1990s by the new, high technology Polymerase Chain Reaction (PCR) test, often known as the viral load test. PCR was not invented as a test, but as a method of manufacturing DNA through an exponential replication process. This leads to problems when it is used as a test because errors also multiply exponentially and the incredible sensitivity of this manufacturing technique results in serious problems with contamination. However, despite these drawbacks, the test has become very popular for HIV. Patients can readily grasp the simple logic “High viral load – bad; Low viral load – good” and are supposed to focus on that (and on their CD4 cell counts) instead of how they feel, which is not considered a reliable indicator of health in the context of AIDS.
It is important to make the results of viral load tests concordant with antibody tests. This can be done by blaming laboratory contamination or by claiming that viral load measures defective, non-infectious virus particles vastly more (by a factor of tens of thousands more) than infectious particles , . In other cases true false positives have to be admitted , , , , , , and false results due to immunization or acute illness , , as well as significant inconsistencies between different brands or types of viral load tests .
The other major type of HIV test is culturing. This is often referred to as isolation. Probably if this test was not so time consuming and expensive and did not produce negative results so often, that it would be the accepted gold standard for the detection of HIV. To call the process of adding unpurified materials to a foreign cell culture (usually immortal cancerous cells), stimulating it with a variety of chemicals (including those that are quite toxic, and including atmospheric oxygen, present at much higher levels than in natural cell systems ) and then, without purifying the resulting materials, claiming that non-specific phenomena such as the detection of reverse transcriptase, HIV-like particles under an electron microscope or p24 constitutes ‘isolation’ is to illustrate that one word can have just as divergent meanings as the word fruit. In fact, the use of the word isolation in the context of HIV culturing is not defined in any dictionary (at least not in the OED) unlike ‘fruit’ which has the scientific and common definitions clearly defined. All definitions of isolation contain the core idea of separating one thing from everything else, something that does not occur anywhere in the process of culturing HIV.
The problem with all of these test systems is that they are all too indirect to constitute proof of the presence of HIV. Yet this fundamental limitation is generally ignored. The only reasonable ‘gold standard’ is purification of HIV from people with AIDS, and the lack of purification from those without it. Purification must be validated by the morphology visible under the electron microscope (size, shape and structure) as well as by analysis of its constituent proteins and RNA. This analysis cannot be performed on impure materials or those of unknown provenance, such as all reagents used today.
Purification according to reasonable standards has never been achieved with HIV. The closest that has been achieved to electron micrograph evidence of purification was not published until 1997 , and this research, based on centrifuged material from cell cultures, indicated that only a minorityof the material that had often previously been called “purified HIV” could possibly be the virus, as most differed in size and shape.
Science demands precision in wording. Popularized science is often characterized by a fast and loose attitude towards definitions. This is found not only in schools, but also throughout medicine. Here the need to obfuscate is not driven by a desire to simplify for an immature audience, but by financial incentives. If “AIDS” is accepted as a term it implies that medicines approved for use in America will be just as useful in Africa. Luckily for the multi-billion dollar industry that has grown up around HIV and AIDS most people are perfectly happy to use these terms to delude themselves into thinking that they understand.
During the writing of this article I asked my Grade 7 daughter Julia whether she knew whether a tomato is a fruit or a vegetable. “Oh yes,” she said “it is a vegetable, because my science teacher told me.” I asked how she defined the words ‘fruit’ and ‘vegetable’ and she was unable to answer. She kept retreating to “My teacher knows” as the ultimate answer, just as we might rely on a fact sheet from the CDC to assure us that HIV is the sole cause of AIDS.
While this is hardly a scientific experiment, it does show that an educational opportunity has been missed, in favour of repetition of a legend. That rote learning should be imposed in the name of grade school science is bad enough, but the consequences for medicine are truly catastrophic.
The only solution is to demand precision. Calling people to task who ask if a tomato is a fruit is a good start, although this discussion will only open minds, not save lives. More importantly, every time the words AIDS or HIV are used in a discussion be sure and ask for them to be precisely defined. Monitor people using these terms, even after the definition has been qualified, for segues from US-AIDS-1993 to WHO-AIDS. Do not let them get away with their assumption that everyone still accepts their grade school teacher’s admonition that “A tomato is a fruit (dummy)”.
1. Oxford English Dictionary on CD ROM for Apple Macintosh. 1993.
3. CDC. Revision of the CDC Surveillance Case Definition for Acquired Immunodeficiency Syndrome. MMWR. 1987 Aug 14; 36(1S): 1S-15S.
4. Revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR. 1992; 41(RR-17): 1-17.
5. HIV/AIDS Surveillance Report (through December 1997). CDC. 1998; 9(2).
6. HIV/AIDS Surveillance Report; U.S. HIV and AIDS cases reported through December 1998. CDC. 1999; 10(2).
7. CDC, Crowe D, Koliadin V. Exchange of correspondence, available upon request.
8. WHO case definitions for AIDS surveillance in adults and adolescents. WER. 1994 Sep 16; 69(37): 273-5.
9. Blanche S. Relation of the course of HIV infection in children to the severity of the disease in their mothers at delivery. N Engl J Med. 1994 Feb 3; 330(5): 308-12.
10. Coombs RW et al. Plasma viremia in human immunodeficiency virus infection. N Engl J Med. 1989 Dec 14; 321(24): 1626-31.
11. Semple M et al. Direct measurement of viraemia in patients infected with HIV-1 and its relationship to disease progression and zidovudine therapy. J Med Virol. 1991; 35: 38-45.
12. Rogers MF et al. Use of the polymerase chain reaction for early detection of the proviral sequences of human immunodeficiency virus in infants born to seropositive mothers. N Engl J Med. 1989 Jun 22; 320(25): 1649-54.
13. McKinney RE et al. A multicenter trial of oral zidovudine in children with advanced human immunodeficiency virus disease. N Engl J Med. 1991 Apr 11; 324(15): 1018-25.
14. Fischl MA et al. A randomized controlled trial of a reduced daily dose of Zidovudine in patients with the Acquired Immunodeficiency Syndrome. N Engl J Med. 1990; 323(15): 1009-14.
15. Faulk WP et al. HIV proteins in normal human placentae. American Journal of Reproductive Immunology. 1991; 25: 99-104.
16. US Public Health Service guidelines for testing and counseling blood and plasma donors for Human Immunodeficiency Virus Type 1 antigen. MMWR. 1996 Mar 1; 45(RR-2).
17. Frenkel LM et al. Genetic Evaluation of Suspected Cases of Transient HIV-1 Infection of Infants. Science. 15 May 1998; 280: 1073-1077.
18. Piatak M Jr et al. High levels of HIV-1 in plasma during all stages of infection determined by competitive PCR. Science. 1993 Mar 19; 259: 1749-54.
19. Onuigbo MAC. Residual HIV-1 RNA After Highly Active Antiretroviral Therapy. JAMA. 2000 Mar 1; 283(9): 301-2.
20. Rich JD et al. Misdiagnosis of HIV Infection by HIV-1 Plasma Viral Load Testing: A Case Series. Ann Intern Med. 1999 Jan 5; 130(1): 37-9.
21. de Mendoza C et al. False positives for HIV using commercial viral load quantification assays. AIDS. 1998; 12(15): 2076-7.
22. More D et al. Utility of an HIV-1 RNA assay in the diagnosis of acute retroviral syndrome. S Med J. 2000 Oct; 93(10): 1004-6.
23. Erice A et al. Performance characteristics of the bDNA 3.0 assy for quantitation of HIV-1 RNA in plasma [abstract]. 7th Conf. Retroviruses and Opp Infections. 2000 Jan 30-Feb 2.
24. Owens DK et al. Polymerase chain reaction for the diagnosis of HIV infection in adults. A meta-analysis with recommendations for clinical practice and study design. Ann Intern Med. 1996; 124: 803-15.
25. Hecht FM et al. Use of laboratory tests and clinical symptoms for identification of primary HIV infection. AIDS. 2002 May 24; 16(8): 1119-29.
26. De Rossi A et al. Antigen detection, virus culture, polymerase chain reaction, and in vitro antibody production in the diagnosis of vertically transmitted HIV-1 infection. AIDS. 1991 Jan; 5(1): 15-20.
27. Saag MS et al. HIV viral load markers in clinical practice. Nat Med. 1996 Jun; 2(6): 625-9.
28. Staprans SI et al. Activation of virus replication after vaccination of HIV-1 infected individuals. J Exp Med. 1995 Dec 1; 182(6): 1727-37.
29. Moss WJ et al. Suppression of Human Immunodeficiency Virus Replication during Acute Measles. J Infect Dis. 2002 Mar 25; 184.
30. Guidelines for Laboratory Test Result Reporting of Human Immunodeficiency Virus Type 1 Ribonucleic Acid Determination. MMWR. 2001 Nov 16; 50(RR20): 1-12.
31. Roy S et al. Oxygen sensing by primary cardiac fibroblasts a key role of p24. Circ Res. 2003; 92.
32. Gluschankof P et al. Cell membrane vesicles are a major contaminant of gradient-enriched human immunodeficiency virus type-1 preparations. Virology. 1997; 230(1): 125-133.
33. Bess JW et al. Microvesicles Are a Source of Contaminating Cellular Proteins Found in Purified HIV-1 Preparations. Virology. 1997; 230(1): 134-144.
Copyright © David Crowe,