Chapter VIII. Instinct
Charles Darwin · 1872. Read the original ↗
Instincts vary and are inherited, so natural selection can accumulate slight, useful changes of instinct into the most complex behaviours, including the cuckoo's parasitism, slave-making in ants, the hive-bee's cells and the sterile castes of social insects.
The chapter pairs well-observed behaviour (heritable instinct variation in domestic animals, Darwin's own wax experiments, gradations within ant nests) with inferences about historical pathways that are plausible but not tested. Its candour about difficulties and its use of risky predictions are real strengths. The step from slight variation to any complex instinct rests largely on 'I can see no difficulty', and several evolutionary scenarios are supposed rather than measured.
Claims, graded
- C1ObservedB
Behavioural dispositions vary between individuals and are inherited. This is shown in domestic animals (dogs, cats, pigeons, fowls), and variation in wild animals is asserted more than documented. Selection of spontaneous variations, working alongside habit, has modified domestic instincts.
“It is notorious how much domestic animals vary in their mental qualities.”
Evidence offered Cats with inherited preferences for different prey, pointers pointing untaught, effects of crosses persisting for generations, tumbler-pigeons, loss of broodiness in fowls. For wild animals he gives a few examples (migration, nests, fear of man) and says space prevents more.
- 1Observation//Inherited variation in domestic behaviour is directly reported, some of it first-hand, but wild variation is mostly asserted.
- 2Hypothesis//The claim could fail, since instincts might not vary or be heritable, though 'mental qualities' is loosely defined.
- 3Prediction/The cases are cited to illustrate the claim, and no risky prediction is set out beforehand.
- 4Experiment//Crosses between breeds act as informal genetic tests, though there are no controlled comparisons.
- 5Replication//Many breeders and observers (Huber, St. John, Audubon, Le Roy) are cited, though they are not independent tests of one design.
- 6Falsifiability//He separates habit from selection using cases such as tumbling pigeons and tame rabbits.
What would move it up Controlled breeding or cross-fostering experiments that measure the heritability of specific behaviours, and documented variation in wild populations for the instincts discussed.
- C2ExtrapolatedD
Because instincts vary and are inherited, natural selection can preserve and accumulate slight, useful variations of instinct to any extent that is profitable. This is how all the most complex instincts originated, and gradations found in related species (collateral lines) corroborate it.
“I can see no difficulty in natural selection preserving and continually accumulating variations of instinct to any extent that was profitable.”
Evidence offered Small heritable changes in domestic breeds and wild nests, plus graded series of instincts among living species (Molothrus, ostrich, parasitic bees, Formica species). He admits no instinct is known from extinct species and that his wild examples are given without detail.
- 1Observation/Slight variation and living gradations are observed, but accumulation into complex instincts is not.
- 2Hypothesis//It predicts gradations in collateral lines and instincts useful to their owners, which could fail.
- 3Prediction//Darwin says he expected to find gradations and was surprised how generally he did, which is a real if loosely specified success.
- 4Experiment/Domestic selection is a partial analogue, but accumulation to complex wild instincts has not been tested.
- 5Replication/Several observers report similar gradations, but nobody has reproduced the accumulation step.
- 6Falsifiability//He names what would count against it (instincts made solely for another's good, no gradations) and considers inherited habit as a rival.
What would move it up Measured fitness differences and heritability for intermediate instincts in living lineages, plus an independent phylogenetic ordering showing the graded series reflects descent. Selection experiments producing a new multi-part instinct would also help.
- C3ObservedB
The hive-bee's hexagonal comb follows from a few simple instincts that bees can be seen to use: excavating equal spherical hollows at a set spacing, stopping where the hollows would intersect, and working on the planes between them.
“all this beautiful work can be shown, I think, to follow from a few simple instincts.”
Evidence offered Darwin's wax-strip and vermilion-ridge manipulations, in which bees dug basins that met at flat walls and stopped at intersection planes. He also gives Miller's geometry of equal spheres, wall-thickness measurements, and Wyman's finding that the real cells are less exact than supposed.
- 1Observation///Darwin watched and manipulated the building behaviour directly and measured wall thickness.
- 2Hypothesis///The geometric rule (spheres at radius times 1.414 spacing yielding the hive-bee's angles) is precise and checkable.
- 3Prediction//He says the theory 'can be tested by experiment', and the thin-ridge result fits what the sphere rule implies.
- 4Experiment//The manipulations are genuine but small, unreplicated within the text, and lack formal controls.
- 5Replication//He follows Tegetmeier's earlier procedure and cites Huber, but the results are not independently repeated here.
- 6Falsifiability//He weighs Huber's rival account and wasp hexagons, and concedes the cells are less precise than claimed.
What would move it up Repeating the wax-manipulation experiments with blind scoring, larger numbers and quantified geometry, and showing that artificial spacing changes predictably change cell shape.
- C4InferredC
The hive-bee's cell-making instinct arose by natural selection acting on successive slight modifications of simpler instincts. The living series of humble-bee, Melipona and hive-bee shows how, and the saving of wax and labour was the advantage that drove it.
“I believe that the hive-bee has acquired, through natural selection, her inimitable architectural powers.”
Evidence offered The living series of cell types, Melipona's near-intersecting spheres joined by flat walls, and the measured cost of wax (twelve to fifteen pounds of sugar per pound of wax, reported by Tegetmeier). The advantage at each step is argued by supposition ('let us suppose').
- 1Observation/Living species and the wax cost are observed, but the ancestral steps and their fitness effects are not.
- 2Hypothesis//The proposed intermediate stages and their economies are specific enough to be wrong.
- 3Prediction/He does not predict an outcome that could have failed; the series was assembled from known species.
- 4Experiment—No test of fitness differences between cell types is reported.
- 5Replication/Huber's and Darwin's descriptions of Melipona agree, but nothing tests the evolutionary step.
- 6Falsifiability/He notes that success may depend on enemies or other causes, but does not say what result would count against the scenario.
What would move it up Evidence that colonies with closer, more regular cells have higher survival and reproduction, together with variation in cell spacing within a Melipona-like species, and a phylogeny showing the living series reflects descent.
- C5InferredC
Sterile insect castes, even those differing widely from their fertile parents, can be explained by selection acting on the family or colony. Colonies whose fertile parents produce useful neuters survive, and gradations among neuters within nests make the route plausible.
“selection may be applied to the family, as well as to the individual, and may thus gain the desired end.”
Evidence offered The analogy with breeders selecting families for marbled meat or long-horned oxen, and Verlot's double-flowered stocks. Also dissection and measurement of gradations among workers in single nests (Formica flava, the driver ant Anomma), with Lubbock's drawings.
- 1Observation//The castes and the gradations within nests are directly examined, but colony-level selection itself is not observed.
- 2Hypothesis//It predicts intermediate forms within nests and heritable tendencies to produce particular castes, both checkable.
- 3Prediction//Darwin says he 'confidently' expected gradations in Anomma and found them, a risky test.
- 4Experiment/Only horticultural and cattle breeding are offered, not manipulation of insect colonies.
- 5Replication/Smith and Bates are cited for analogous cases, but the colony-level step is not independently tested.
- 6Falsifiability//He calls the neuters 'by far the most serious special difficulty', states why they could be fatal to the theory, and tests the gradation prediction.
What would move it up Measured heritable variation among colonies in the proportion or form of castes, and measured differences in colony survival and reproduction that track those differences.
- C6InferredC
Each species' instincts are good for the species itself and have never been produced for the exclusive good of another species. Instincts are also not always perfect, as the theory predicts.
“no instinct can be shown to have been produced for the good of other animals”
Evidence offered A test with aphids and ants, in which Darwin excluded ants, tickled the aphids with a hair without result, then let an ant stroke them and saw them excrete. He infers that excretion also benefits the aphid, and cites imperfect instincts (Molothrus, ostrich eggs strewn over the plains).
- 1Observation//The aphid manipulation and the wasted eggs are directly observed, but 'benefit' is inferred.
- 2Hypothesis///A single clear case of an instinct existing only to benefit another species would count against the claim.
- 3Prediction//It predicts that supposed altruistic instincts will turn out to benefit their owners, and the aphid case fits.
- 4Experiment//Excluding ants and testing a hair stimulus is a simple controlled comparison on one species.
- 5Replication/Huber's earlier observation agrees, but the general claim rests on few cases.
- 6Falsifiability///The claim names a kind of result that would refute it.
What would move it up A systematic survey of putatively other-serving instincts, with measured costs and benefits to the performer (for example, how much aphids gain from removal of the excretion).
Reasoning check
- Clear caseClaim C2
“I can see no difficulty in natural selection preserving and continually accumulating variations of instinct to any extent that was profitable.”
The observed cases are slight changes (pointing, retrieving, nest differences), yet the conclusion is that complex instincts such as the cuckoo's ejection of its foster-brothers were built the same way. The step from small to large rests on 'I can see no difficulty' and on a deferral ('want of space prevents me'), though Darwin also offers living gradations as support.
- PossibleThe argument as a whole
“is plainly explicable on the foregoing views, but is otherwise inexplicable”
Imperfect instincts, instincts not made for others' good and the absence of abrupt leaps are counted as corroboration only if the rival view (specially created instincts) would predict the opposite. That premise about what a Creator would do is never stated or argued. The real evidence offered is the documented imperfections (Molothrus, the ostrich's strewn eggs), and a stated, independently motivated account of what the rival predicts would settle the point.
What it does well
- Darwin openly names the neuter insects as 'by far the most serious special difficulty', and he reports facts that cut against his own tidy picture (Wyman's finding that bee cells are less exact than supposed, and imperfect instincts).
- He makes a prediction that could fail (graded intermediates among castes in a single nest) and reports the test with dissections and drawings.
- The bee-cell section includes simple experiments with manipulable variables and a precise geometric model checked by a mathematician.
- The source of variation is named separately from selection ('spontaneous variations... produced by the same unknown causes'), and advantage is often defined by independent measures such as the sugar cost of wax, not by survival alone.
- He separates habit and selection as causes, uses neuter insects as a test against Lamarckian inheritance, and states that the chapter does not greatly strengthen his theory.
Limits of this evaluation This is one chapter of a longer book and Darwin often says he lacks space for details, so some support may lie elsewhere. The focus checks were applied closely. Reification: 'natural selection led the bees' and 'nature reveals her method' are shorthand, since the mechanism (heritable variation, differential success of swarms or colonies, struggle for existence) is spelled out, and the argument survives rewriting without agent words, so no finding is listed. Darwin also states that the bees do not know what they are doing. False analogy: breeder comparisons are used to show that family-level selection and heritable change are possible, and he supplies the natural counterpart (survival of whole communities), so the conclusion does not depend on breeder foresight. Tautology: advantage is generally defined independently (wax and labour saved, incubation time, strength), though in the cuckoo and ant scenarios advantage is only supposed. The closing phrase 'let the strongest live and the weakest die' is a rhetorical summary, which Darwin flags as not a logical deduction. Filter as source: variation is named as the source and shown for many traits, but not for each specific novel component, which is covered under C2. Later work (heritability studies of behaviour, inclusive-fitness theory for colonies) bears on C1 and C5 and supports their direction, but was not available to Darwin.