| 10952 | Unusual kinds like mule are just a combination of two kinds [Aristotle] |
| Full Idea: The kind that is common to both horse and ass and which most nearly comprises them happens not to have a name, but can safely be presumed to be both, i.e. the horse-ass or 'mule'. ...A mule does not come from a mule. | |
| From: Aristotle (Metaphysics [c.324 BCE], 1033b32) | |
| A reaction: [second part at 1034b04] Does ancient Greek have a word for 'mule' - it sounds as if it doesn't. Nice chicken-and-egg problem. Must a natural kind be derived from a natural kind? No. Gold does not derive from gold. |
| 19244 | Chemists rely on a single experiment to establish a fact; repetition is pointless [Peirce] |
| Full Idea: The chemist contents himself with a single experiment to establish any qualitative fact, because he knows there is such a uniformity in the behavior of chemical bodies that another experiment would be a mere repetition of the first in every respect. | |
| From: Charles Sanders Peirce (Reasoning and the Logic of Things [1898], IV) | |
| A reaction: I take it this endorses my 'Upanishads' view of natural kinds - that for each strict natural kind, if you've seen one you've them all. This seems to fit atoms and molecules, but only roughly fits tigers. |
| 7375 | Quine probably regrets natural kinds now being treated as essences [Quine, by Dennett] |
| Full Idea: The concept of natural kinds was reintroduced by Quine, who may now regret the way it has become a stand-in for the dubious but covertly popular concept of essences. | |
| From: report of Willard Quine (Natural Kinds [1969]) by Daniel C. Dennett - Consciousness Explained 12.2 n2 | |
| A reaction: He is right that Quine would regret it, and he is right that we can't assume that there are necessary essences just because there seem to be stable natural kinds, but personally I am an essentialist, so I'm not that bothered. |
| 16935 | If similarity has no degrees, kinds cannot be contained within one another [Quine] |
| Full Idea: If similarity has no degrees there is no containing of kinds within broader kinds. If colored things are a kind, they are similar, but red things are too narrow for a kind. If red things are a kind, colored things are not similar, and it's too broad. | |
| From: Willard Quine (Natural Kinds [1969], p.118) | |
| A reaction: [compressed] I'm on Quine's side with this. We glibly talk of 'kinds', but the criteria for sorting things into kinds seems to be a mess. Quine goes on to offer a better account than the (diadic, yes-no) one rejected here. |
| 16936 | Comparative similarity allows the kind 'colored' to contain the kind 'red' [Quine] |
| Full Idea: With the triadic relation of comparative similarity, kinds can contain one another, as well as overlapping. Red and colored things can both count as kinds. Colored things all resemble one another, even though less than red things do. | |
| From: Willard Quine (Natural Kinds [1969], p.119) | |
| A reaction: [compressed] Quine claims that comparative similarity is necessary for kinds - that there be some 'foil' in a similarity - that A is more like C than B is. |
| 6613 | The natural kinds are objects, processes and properties/relations [Ellis] |
| Full Idea: There are three hierarchies of natural kinds: objects or substances (substantive universals), events or processes (dynamic universals), and properties or relations (tropic universals). | |
| From: Brian Ellis (Katzav on limitations of dispositions [2005], 91) | |
| A reaction: Most interesting here is the identifying of natural kinds with universals, making universals into the families of nature. Universals are high-level sets of natural kinds. To grasp universals you must see patterns, and infer the underlying order. |
| 5472 | Natural kinds are of objects/substances, or events/processes, or intrinsic natures [Ellis] |
| Full Idea: Natural kinds appear to be of objects or substances, or of events or processes, or of the intrinsic nature of things; hence there should be laws of nature specific to each of these categories. | |
| From: Brian Ellis (The Philosophy of Nature: new essentialism [2002], Ch.5) | |
| A reaction: It is nice to see someone actually discussing what sort of natural kinds there are, instead of getting bogged down in how natural kinds terms get their meaning or reference. Ellis recognises that 'intrinsic nature' needs some discussion. |
| 15886 | Science rests on the principle that nature is a hierarchy of natural kinds [Harré] |
| Full Idea: The animating principle behind the material and discursive practices of science is the thesis that nature exemplifies multiple hierarchies of natural kinds. | |
| From: Rom Harré (Laws of Nature [1993], 5) | |
| A reaction: I agree. I take it to be a brute fact that there seem to be lots of stable natural kinds, which are worth investigating as long as they stay stable. If they are unstable, there needs to be something stable to measure that by - or we give up. |
| 17371 | Some kinds are very explanatory, but others less so, and some not at all [Devitt] |
| Full Idea: Explanatory significance, hence naturalness, comes in degrees: positing some kinds may be very explanatory, positing others, only a little bit explanatory, positing others still, not explanatory at all. | |
| From: Michael Devitt (Natural Kinds and Biological Realism [2009], 4) | |
| A reaction: He mentions 'cousin' as a natural kind that is not very explanatory of anything. It interests us as humans, but not at all in other animals, it seems. ...Nice thought, though, that two squirrels might be cousins... |
| 17381 | Phylogenetics involves history, and cladism rests species on splits in lineage [Dupré] |
| Full Idea: The phylogenetic conception of classification reflects the facts of evolutionary history. Cladism insists that every taxonomic distinction should reflect an evolutionary event of lineage bifurcation. | |
| From: John Dupré (The Disorder of Things [1993], 1) | |
| A reaction: Devitt attacks cladism nicely. It rules out species change without bifurcation, and it insists on species change even in a line which remains unchanged after a split. |
| 17385 | Kinds don't do anything (including evolve) because they are abstract [Dupré] |
| Full Idea: A kind, being an abstract object, cannot do anything, including evolve. | |
| From: John Dupré (The Disorder of Things [1993], 2) | |
| A reaction: Maybe. We might have an extensional view of the kind, so that 'gold' is the set of extant gold atoms. But possible gold atoms are also gold, and defunct ones too. Virtually every word in English is abtract if you think about it long enough. |
| 6768 | Tin is not one natural kind, but appears to be 21, depending on isotope [Bird] |
| Full Idea: If real essences are decided by microstructure, then what we call the element tin is not a natural kind, but a mixture of 21 different kinds, one for each isotope. There also exist two different allotropes of tin - white tin and grey tin. | |
| From: Alexander Bird (Philosophy of Science [1998], Ch.3) | |
| A reaction: This example vividly brings out the difficulties of the Kripke-Putnam view. If natural kinds 'overlap', then there would be a very extensive overlap among the 21 isotopes of tin. |
| 6771 | Natural kinds may overlap, or be sub-kinds of one another [Bird] |
| Full Idea: It seems clear that in some cases one natural kind may be a subkind of another, while in other cases natural kinds may overlap without one being the subkind of another. | |
| From: Alexander Bird (Philosophy of Science [1998], Ch.3) | |
| A reaction: Given the enormous difficulty of pinpointing natural kinds (e.g. Idea 6768), it is hard to know whether the comment is correct or not. Ellis says natural kinds come 'in hierarchies', which would make subkinds normal, but overlapping unlikely. |
| 6770 | Membership of a purely random collection cannot be used as an explanation [Bird] |
| Full Idea: One might randomly collect diverse things and give the collection a name, but one would not expect it to explain anything to say that a certain object belonged to this collection. | |
| From: Alexander Bird (Philosophy of Science [1998], Ch.3) | |
| A reaction: This is in support of Bird's view that natural kinds are formulated because of their explanatory role. There is, though, an undeniable subjective aspect to explanation, in that explanations arise from the ignorance and interests of persons. |
| 6776 | Natural kinds are those that we use in induction [Bird] |
| Full Idea: Natural kinds are the kinds one should make use of in inductive inference (if that is explanation which leads to laws). | |
| From: Alexander Bird (Philosophy of Science [1998], Ch.3) | |
| A reaction: The problem with this is that it is epistemological rather than ontological. In induction we use superficial resemblences that are immediately obvious, whereas the nature of kinds can be buried deep in the chemistry or physics. |
| 6767 | Rubies and sapphires are both corundum, with traces of metals varying their colours [Bird] |
| Full Idea: Both rubies (valuable) and sapphires (less valuable) are corundum (Al2O3), differing only in their colours, for which traces of iron, titanium and chomium are responsible. | |
| From: Alexander Bird (Philosophy of Science [1998], Ch.3) | |
| A reaction: A nice example which illustrates how natural kinds determined by nominal essence could be drastically different from those suggested by real essence. It certainly suggests that corundum might be a natural kind, but ruby isn't. |
| 17394 | Natural kinds are what are differentiated by nature, and not just by us [Scerri] |
| Full Idea: Natural kinds are realistic scientific entities that are differentiated by nature itself rather than by our human attempts at classification. | |
| From: Eric R. Scerri (The Periodic Table [2007], Intro 'Evol') |
| 17421 | If elements are natural kinds, might the groups of the periodic table also be natural kinds? [Scerri] |
| Full Idea: Elements defined by their atomic numbers are frequently assumed to represent 'natural kinds' in chemistry. ...The question arises as to whether groups of elements appearing in the periodic table might also represent natural kinds. | |
| From: Eric R. Scerri (The Periodic Table [2007], 10 'Elements') | |
| A reaction: Scerri says the distinction is not as sharp as that between the elements. As a realist, he believes there is 'one ideal periodic classification', which would then make the periods into kinds. |
| 14504 | The Kripke/Putnam approach to natural kind terms seems to give them excessive stability [Koslicki] |
| Full Idea: Theoretical terms such as 'mass', 'force', 'motion', 'species' and 'phlogiston' seem to indicate that the Kripke/Putnam approach to natural kind terms is committed to an excessive amount of stability in the meaning and reference of such expressions. | |
| From: Kathrin Koslicki (The Structure of Objects [2008], 8.6.2) | |
| A reaction: This sounds right to me. The notion of 'rigid' designation gives a nice framework for modal logic, but it doesn't seem to fit the shifting patterns of scientific thought. |
| 18613 | Artifacts can be natural kinds, when they are the object of historical enquiry [Machery] |
| Full Idea: Some artifacts are the objects of inquiry in the social sciences ...such as prehistoric tools ...and hence, artifacts are bona fide natural kinds. | |
| From: Edouard Machery (Doing Without Concepts [2009], 8.2.1) | |
| A reaction: Presumably if a bird's nest can be a natural kind, then so can a flint axe, but then so can a mobile phone, for an urban anthropologist. 'Natural' is, to put it mildly, a tricky word. |