The nature of wolf packs is crucial to understanding the social behaviour of wolves, and thus the behavioural inheritance of domestic dogs, but until recently wolf packs have been wrongly thought of as competitive organizations. It is now known that the majority of wolf ‘packs’ are simply family groups. Typically, a solitary male will pair up with a solitary female – either or both will most likely have recently left a pack – and raise a litter together. In many species the young leave or are chased away when they are old enough to fend for themselves, but not so in wolves. Provided that no one is starving, the cubs may stay with their parents until they are fully grown. Once they are experienced enough, they will participate fully in hunting, and thus a pack emerges. Often the younger members will still be part of the pack when the next litter of cubs is born, and will help their parents to raise their brothers and sisters, bringing food back for them and babysitting them when the other members of the pack are out hunting. Contrary to many notions of wolf behaviour, co-operation, not dominance, seems to be the essence of the wolf pack.
Modern biology demands explanations for the seemingly selfless behaviour exhibited by the younger adult wolves operating within a pack. Logically speaking, a gene that influences one animal to help another to breed should die out, since it will be the animals not carrying the ‘unselfish’ version of the gene that will leave the most offspring. Co-operative breeding must therefore have long-term advantages that outweigh its disadvantages. Biologists have been arguing for the past five decades as to what the finer points of those advantages might be and how they are expressed, but the theory of kin selection, first proposed in the 1960s, robustly accounts for why co-operative breeding is much more likely to occur in families than in random assemblages of individuals.
By ascribing a benefit to the co-operation observed within wolf packs, scientists have been able to use the theory of kin selection to make sense of behaviour that would otherwise be unintelligible. Offering co-operation to an unrelated animal carries risks, even in animals as smart as wolves; there is a danger, after all, that the favour may not be returned. On the other hand, performing a favour for a close relative, say a son or a daughter, has genetic advantages; even if that favour is never returned, the helper is still promoting the survival of some of his or her own genes, specifically those versions that are identical in their relative. (In a son or daughter there will be a 50 per cent overlap, the other half coming from the other parent.) This advantage does not seem sufficient to promote lifelong abstinence from breeding – the only mammals which exhibit such abstinence are naked mole-rats, which burrow beneath harsh deserts where a single breeding pair is unlikely to survive for long, even with the others’ help. Nevertheless, kin selection does seem to be powerful enough to sustain temporary abdication from breeding, making it worthwhile for offspring to help their parents until the family group gets too large to sustain itself, and the offspring leave to start families of their own.
Kin selection explains that, when the younger members of the pack appear to deliberately put their own breeding rights to one side, they are actually acting in their own interest – but the advantages of this behaviour are not purely familial. In addition to the advantage they gain from kin selection, it is also safer for the wolves themselves not to leave their pack while they are still young. Their lack of experience means that their chances of forming their own pack are actually rather slim. This accounts for the rare occasions when unrelated wolves have been recorded as joining existing packs; they often seem to be recruited as a replacement when one of the most experienced members of the pack, maybe one of the original founders, leaves or dies.
Packs that form naturally, in the wild, are usually harmonious entities, with aggression being the exception rather than the norm. As in any family, there are occasional conflicts of interest within wolf packs, but in general the parents have to do very little to keep their grown-up young in order. The young are essentially volunteers – they could leave the family and set up on their own, but they choose not to, preferring to stay safely within the family unit until they are older and more experienced, and hence more likely to survive the risks of finding a mate and a new place to live. They regularly reinforce their bond with their parents, and at the same time reassure them that they are helpers, not rivals, by performing a special ritual. The youngster crouches slightly as it approaches the parent, ears back and close to its head, and tail held low and wagging. It then nuzzles the side of the parent’s face, an imitation of the food-soliciting behaviour that it used when it was a cub. (This is very similar to the greeting ritual of the African wild dog, and so possibly a very ancient canid behaviour, predating the evolution of both wolf and wild dog.)
The image of a harmonious pack is not the picture of wolf society that you will find in most books on dog behaviour. Wolf biologists originally based most of their ideas on captive packs, which were easy to observe. Some of these packs were random assemblies of unrelated individuals, while others were fragments of packs, usually with one or both of the parents missing – basically composed of whatever individuals were available for the zoo to make an exhibit. What almost all of these packs had in common was that their structure had been irrevocably disrupted by captivity, so that the wolves were thrown into a state of confusion and conflict. Moreover, unless their human captors decided to separate them, none had the opportunity to leave. As a result, the relationships that emerged were based not on long-established trust, but on rivalry and aggression.
The true picture of wolf society emerged only as wolves became protected, allowing packs to form and thrive over several years, without becoming fragmented by continual persecution. At roughly the same time, better technology for tracking and observing wolves in the wild became available: GPS, miniature radio transmitters with batteries robust enough to allow tracking over a whole season, and so on. Within a decade, descriptions of wolf society had changed from the image of the hierarchical pack run by two tyrants, one male, one female, to that of the harmonious family group, where, barring accidents, the younger adults in the family voluntarily assisted their parents in raising their younger brothers and sisters. Coercion was replaced by co-operation as the underlying principle.
This radical change in our conception of pack behaviour has required that we also reappraise the social signals that wolves use. Under zoo conditions, signals that wolf parents would normally use to remind their offspring to co-operate instead became the precursors of out-and-out fighting, and were labelled ‘dominance indicators’. Similarly, the cohesive behaviours that the adult young wolves would normally use to bond with their parents were now being used in desperate attempts to avoid conflict, and so came to be labelled as ‘submission’.
Contrary to long-standing theories of wolf behaviour, it is now believed that ‘submissive’ behaviour may be nothing of the sort. An effective ‘submissive’ display should, by definition, indicate to an attacker that the attack is not worth pursuing – and, indeed, when wolves from two different packs happen to meet, the smaller one will try to avoid being attacked by performing such a display. This rarely works, however, and if the smaller wolf fails to run away, it will be attacked and often killed by the larger. Wolves from different packs have no common interests; they compete for food and are probably only very distantly related, if at all. Nevertheless, if the ‘submissive’ display was truly an indication of submission, it ought to work in these circumstances, since the attacking wolf is putting itself at risk of injury, even if it wins. The fact that this display does not work under these circumstances indicates that it is not a ‘submissive’ display at all. Moreover, when it is performed between members of the same family, for the most part it is not preceded by any form of threat from the recipient. Rather, it usually appears spontaneously, reinforcing the bond between the members of the pack. Only in artificially constituted ‘packs’, kept in zoos, do ‘submissive’ displays come to be a standard response to threat. Presumably the younger, weaker wolves learn by trial
and error that such displays (sometimes) work under these unnatural circumstances, where pack loyalties have been totally disrupted, and there is nowhere for them to escape to.
Wolves perform two signals that used to be labelled ‘submissive’: ‘active’ and ‘passive’. Domestic dogs perform very similar signals, and these, too, are conventionally referred to as ‘active submission’ and ‘passive submission’. One might expect that any reinterpretation of these signals in the wolf would have been quickly followed by a reappraisal of what they mean when performed by dogs, but this has been slow to happen.
A Wolf (left) performing the affiliation display
The ‘active’ display is the more common of the two in wolves, and rather than being a sign of submission, is in fact a bonding signal that scientists now refer to – much more appropriately – as the ‘affiliation display’. In the affiliation display, the wolf approaches with a low posture, holding its tail low; its ears are pulled slightly back, and its tail and hindquarters wag enthusiastically. This display forms part of what is called the ‘group ceremony’, which occurs when the pack reassembles, or as a precursor to a hunting trip. Under these circumstances it can be performed by the parents (the so-called ‘alphas’) as well as by their offspring, confirming its role as a mechanism whereby affectionate bonds are reinforced. It is actually difficult to work out how this display was ever labelled as ‘submissive’ behaviour. A wolf performing the affiliation display is actually in rather a good position to attack its recipient – a swift twist of the head and it could sink its teeth into the other’s throat. Therefore accepting the performance of the affiliation display is, if anything, more of an expression of trust on the recipient’s part than on the performer’s. It is undeniable that the younger members of the family perform the affiliation display towards their parents much more frequently than vice versa, but this is typical of all parent–offspring relationships, and does not mean that the offspring are allowing themselves to be ‘dominated’ by their parents. Indeed, it simply reflects the asymmetry of the relationship between parents and offspring. These are the only parents that the young wolves will ever have, and therefore their attachment is total. The parents may – indeed probably do – have other offspring, so their attachment to each cub must unavoidably be a shared one.
Also in urgent need of reinterpretation is the wolf’s other, less common submissive display, ‘passive submission’. Unlike the affiliation display, this may be an actual sign of submission, one that is derived from an infantile behaviour, in which cubs roll over to allow their mother to groom their belly and stimulate urination, which the cubs cannot yet do on their own. This display, which seems to have been adopted by adults as a way of deflecting possible attack, involves one wolf lying down, rolling on to its back, and exposing its abdomen for inspection by another. Some wolf biologists now refer to it as the ‘belly-up display’, which is more descriptive and presumes nothing about its function. Nevertheless, this display has precisely the characteristics that one would expect of submission, since the wolf that performs it is placing itself in a position where it is at the mercy of the other.
The belly-up display
In wolves, the belly-up display is much rarer than the affiliation display, and is more commonly seen in captive wolves than in the wild. When observed in zoos, it is most likely to be performed by wolves that are on the fringes of the captive ‘packs’, are often involved in fights, and rarely participate in group-howls. These are the very wolves that would almost certainly have gone off on their own if the fence surrounding their enclosure had not been there.5 Continually stressed by being forced to remain in close quarters with other wolves that threaten to attack them at every turn, these outcasts will try any tactic that might deflect aggression. In cases where the affiliation display fails, behaving like a helpless cub evidently works, so they learn to use it when they are in desperation. Thus in wolves this display may be an artefact of captivity – not a normal part of adult behaviour at all, but rather a signal artificially carried over from infancy into adulthood, under unnatural circumstances.
Studies of wild wolf packs have made clear that the traditional interpretations of wolf submission – both aggression within the pack, and the ‘submissive’ behaviour that is an attempt to defuse or deflect this aggression – reflect artefacts of captivity, and therefore cannot be applied to wolves as a species. Wolf packs that have not been manipulated by man, and are allowed to manage their own affairs, so to speak, are generally peaceful. This is not to say that all of what has been written about aggression between wolves is wrong; for instance, it is an undeniable fact that wolves – captive or not – can be very forceful and aggressive when they want to be. In the wild, even though relationships within packs are usually congenial, aggression towards outsiders, though infrequent, is unrestrained and potentially fatal. In captivity, however, pack ‘identity’ is either non-existent or severely disrupted, resulting in the expression of behaviour that would normally be seen only in skirmishes between members of different packs.
Observations of captive wolf packs have led not only to mistaken assessments of wolf behaviour, but also to fundamental misunderstandings about the structure of wolf families themselves, misunderstandings that have warped the popular conception of dogs as well. In captive wolf packs, the breeding pair are conventionally referred to as the ‘alpha male’ and the ‘alpha female’. Many dog trainers, borrowing from this conception, insist that owners must impress their own ‘alpha’ status on their dog, which would otherwise be driven to seek ‘alpha’ status for itself. Thanks to our new understanding of the way that wild wolves construct their packs, however, it has become clear that ‘alpha’ status comes automatically with being a parent. The term ‘alpha’, as applied to a parent wolf in a normal pack, does not thus describe much about the wolf’s status beyond its role as a parent.6 It is meaningful only when used to describe the eventual victor of the warfare that is endemic to captive groups of wolves, which lack the family ties that would ensure peace in a natural pack. Which of these two models points to the most appropriate way to understand pet dogs and their relationships with their owners? Is it the ‘alpha’ model, based upon the unnatural captive pack, or the ‘family’ model, based upon the behaviour of wolves that have been allowed to make their own choices as to who or who not to live with? The family model is the product of millions of years of evolution, allowing the formation and refinement of an elaborate set of signals that serve to keep the peace. The alpha model emerges only in artificial social groupings that evolution has never had the opportunity to act upon, and in which individual wolves have to draw upon every ounce of their intelligence and adaptability just in order to survive the relentless social tensions inherent in such groupings.
Putting aside these objections about popular conceptions of wolf hierarchies for a moment, we should note two further reasons why understanding wolf behaviour cannot be the be-all and end-all for understanding dog behaviour. Scientists have (inadvertently, to be fair) studied the wrong wolves; on the wrong continent, and 10,000 years too late.
The American timber wolf, which is a sub-species of the grey wolf, is the most studied wolf of all, and has long been used to interpret the behaviour of dogs.7 Until very recently, researchers tacitly assumed that the American timber wolf was closely related to the domestic dog, and therefore that studies of American wolves were highly relevant to understanding what makes dogs tick. However, the advent of DNA technology has forced a reappraisal of this comparison between dogs and American timber wolves.
Apart from deliberately bred wolf-dog hybrids, it has not been possible to trace the DNA of any of the dogs in the Americas to North American wolves, not even the ‘native’ dogs that were there before Columbus. This lack of evidence has not been for want of effort. The first such genetic analysis was done on the Mexican hairless dog, or xoloitzcuintli. The Spanish conquistadors discovered this dog when they first arrived in Mexico, where it was used for a variety of purpose
s, including companionship and food; it was also believed to have healing powers. To avoid contamination of these supposed properties through cross-breeding with European dogs, xolos were reportedly bred secretly in isolated locations dotted throughout western Mexico, and their descendants survive to this day. Could these dogs be relics of an ancient domestication of New World wolves? Their mitochondrial DNA (inherited through the female line) proves otherwise: it is most similar to that of European dogs (and wolves), bearing much less resemblance to that of American wolves.
Although the xolo’s genetic makeup suggested that this breed originated in Europe, not South America, the possibility still remained that the American continents had produced other native dogs. Indeed, the modern xolos examined by scientists might conceivably not have belonged to an ancient breed at all, but instead might have been facsimiles re-created by breeders from crosses between European breeds – a possibility that would explain why the modern xolos’ DNA is European in type. So genetic researchers next turned their attention to other native American dogs; to DNA extracted from the marrow of 1,000-year-old (or older) dog bones taken from archaeological digs in Mexico, Peru and Bolivia, and also from the bones of dogs buried in the permafrost of Alaska prior to the discovery of that area by Europeans in the eighteenth century. In both cases, the DNA was much more similar to that of European wolves than to that of American wolves.
In Defence of Dogs Page 4