Border Collies are the go-to example of a breed that has been honed and improved through a performance and working standard. They are generally healthy and well behaved dogs that are multi-talented and excel at just about any dog endeavor save protection. But the mantra of perfection through trial performance has its limits. Don’t tell that to the trail elitists though. One prominent Border Collie trialist and author (Do Border Collies Dream of Sheep?) thinks that Hip Dysplasia must be advantageous to Border Collies because the supposedly comprehensive selection tool of trial success hasn’t cleansed the breed of Hip Dysplasia in the same manner that selection for fast racers has done so in the sight hound breeds.
Take a moment to read C. Denis Wall, PhD’s argument titled My Thoughts on Developing a Wider View of Hip Dysplasia in the Border Collie and return for my thoughts.
There’s probably a psychological compensating behavior defined by a person making the mental conversion of an intractable problem from a detriment into a supposed advantage. “It’s not a bug; it’s an undocumented feature!” This self-delusion protects the person from having to confront harsh and unwelcome truths which would otherwise damage their self image or world view. I believe that this self-deceiving behavior is consistent with what psychologists call “motivated reasoning.” This happens when we believe what reaffirms our biases and suits our purposes rather than what is supported by facts, is verifiable, and objective.
While we generally accept negative outcomes that are unambiguous, when there’s uncertainty or ignorance involved motivated reasoning occurs when we support interpretations that benefit us the most or reaffirm our extant biases. This is a form of “self-serving attributional bias” where success is internalized and failure is externalized by taking personal credit for positive outcomes and blaming others for negative outcomes. When an individual has invested a great deal of time or effort, or perhaps sacrificed and even suffered for goal but then learns something new that calls the value of that goal into question, cognitive dissonance will fuel motivated reasoning about the value of that goal: re-evaluating its value more positively than is merited by objective, dispassionate standards.
I believe this is what is happening with C. Denise Wall, PhD’s essay: she doesn’t want to admit that the all-mighty sheep trial selection tool is insufficient to solve hip-dysplasia in Border Collies and instead of being objective she creates a fantasy where a disease must be beneficial because otherwise she’d have to negatively re-evaluate the sheep trial as a necessary, sufficient, and optimal selection tool. Her essay gets interesting when she identifies a piece of new information that calls sheep trials into question:
Then what about Border Collies and PennHIP?
Interestingly, when using strong selection pressure only for their specialized performance, Greyhounds, Salukis and other sight hounds coincidentally develop nearly perfect, very tight hips over generations. Breeding selection based on efficient sled-dog performance is another type of work-only selection that results in good hips. Even when Labrador Retrievers (a breed with a fairly high predisposition for HD), are used as sled dogs, and selected and bred based entirely on their efficiency as workers, they will coincidentally breed good, tight hips over time. Therefore, in these types of performance areas, selection for efficient function alone is strongly associated with incidental selection for good, tight hips.
So why, after all the years of strict selection for herding work, do Border Collies still have hip dysplasia? Why have they not naturally gravitated toward the perfect hip status over all these generations like the racing Greyhound?
These are absolutely the correct and rational questions to ask when presented with the observation that performance selection for running has successfully lowered HD incidence in several breeds but Border Collies show no unique immunity to the disease despite having a clear selection bias based on a performance trait. The problem comes with C. Denise Wall, PhD’s proposed answer to these questions. What do you think a rational person with no investment in sheep trials would propose as the first likely hypothesis for this observation? I contend that you’d call into question the value of the sheep trial as a selection metric for HD. Instead, C. Denise Wall, PhD calls into question the deleterious nature of Hip Dysplasia!
Perhaps, in our breed, one of the risk factors for HD, laxity, is something that has been inadvertently selected for in their performance.
Border Collies are neither better nor worse than the vast majority of other breeds in terms of their incidence of hip dysplasia or their likelihood of developing degenerative joint disease based upon their hip laxity. Their PennHIP distribution curve looks like most other breeds, and although PennHIP has not released laxity vs. Degenerative Joint Disease (DJD) charts for every breed, when I’ve inquired with them they state that most breeds seem to follow the Labrador Retriever curve.
It’s notable that German Shepherds are more likely to develop DJD across all laxity measurements versus all other breeds. Rottweiler’s are moderately more likely to develop DJD at most indexes but slightly less likely to develop DJD at the extreme laxity measures than most other breeds. The Golden Retrievers and Laborador Retrievers appear to conform to the typical risk for DJD versus DI of most breeds.
So the simple and direct hypothesis would be something like “just as selective breeding for field trial performance has not rid the retrieving breeds of hip dysplasia or changed their HD/DJD profile compared to most other breeds, so too has selection via sheep trials failed to change the hip profiles of Border Collies; therefore we contend that success at field trials and sheep trials is not sufficiently harmed by hip dysplasia and breeders selecting stock for success in those events have not altered their choices sufficiently to distinguish their breeds from the average.” Instead, C. Denise Wall, PhD invents a completely unfounded theory that laxity means “flexibility” and is thus a good thing that gives Border Collies a distinct advantage in lateral movement. Utter nonsense, but here’s her pitch:
On the list of breeds that maintain or convert to good hip status when breeding selection is based solely on performance, are breeds that run or pull straight ahead, or mostly so. Border Collies, on the other hand, do a great deal of quick flanking and lateral movement, quick starts and stops, pivots on the hind legs, and traveling in a crouched position in their work. It seems likely that there needs to be some “give” somewhere for efficient performance, and that give has been selected for. In the same way slight cow hocks allow for efficient performance and have been selected for through the many generations of working stock, so perhaps has some laxity in the hips.
The sight hound breeds are hardly limited to simple straight line motion. Any sight hound bred for hunting and coursing can dodge, turn, and juke just fine. Mushing dogs are also tasked with significant lateral movement at speed (lest we are to believe that sledding paths are all straight lines with no curves, well groomed, and level. You might make a note for how track greyhounds are conditioned to run given that formal sport betting racetracks are level and near-straight-line ovals; but this is a matter of their training and conditioning more than significant physical differentiation from the other sight hounds which also share the predisposition against DJD and Hip Dysplasia.
If you take a moment to think about the implications of her hypothesis, C. Denise Wall, PhD’s claim requires you to believe that the Laxity-Advantage in Border Collies would be perfectly balanced with the Hip Dysplasia-Disadvantage because Border Collies are no better or worse than average compared to other breeds.
The Laxity-Advantage would be highly selected for in Border Collies just like their “eye” was highly selected for and thus it must be universal or near-universal in the breed. But Border Collies are just average. If her laxity-as-advantage theory were true, wouldn’t we expect BCs to be plagued by bad hips just like they are ‘plagued’ by “the eye?” What are the odds that we have this pro-laxity selection factor occurring in exact proportion to some anti-HD selection with sheep trial performance and that those two forces which must be strong because sheep trials are a strong selection factor, but miraculously they BALANCE making Border Collies nothing-special-either-way in terms of HD and DJD despite having intense and effective selection tools at work?
If you were double-blind testing a new cancer drug against a placebo and your results showed that the pills you tested had no effect at all on the rate of cancer, would you conclude that (1) You tested the Placebo, (2) Your drug was ineffective but also harmless and thus behaved like the Placebo would, or (3) Your drug both cured cancer in the same rate as it caused the cancer to worsen creating a cosmic wash? This is essentially what Wall wants you to believe laxity’s effect is on the Border Collie and it’s preposterous.
It is an extraordinary claim that lacks extraordinary evidence.
We should also be suspect of this argument on its face because Wall is arguing to maintain the status-quo when presented with information that should disrupt the status-quo. The realization that Border Collie selection culture has not improved laxity and HD has created cognitive dissonance in Wall and instead of decreasing her assessment of the value of that selection culture (sheep trials don’t select for good hips and we do want good hips), she is attempting to reduce the value of the dissonant information (laxity isn’t so bad!).
I ran Wall’s essay by a respected Veterinarian who writes extensively on inherited canine disease and hip dysplasia asking if there was any body of evidence that hip joint laxity conferred an advantage as Wall states. The response was crickets chirping, “she misses the point” regarding PennHIP, and “btw, what’s her PhD in?” Wall’s PhD is in BioChemistry and her BS is in Medical Technology so she should be capable of backing up her extraordinary claim with scientific evidence. The papers she cites at the bottom of her article, however, provide no clarity on the hypothesis that hip laxity is required and/or beneficial for lateral movement and adroit flexibility or agility. The only people making a similar claim are Lundhund fabulists who claim that their breed’s inbred soft tissue issues make them amazing spelunkers and spiderman-like climbers. More motivated reasoning if you ask me.
Wall’s argument gets worse. She goes from denial into attacking hip-testing regimes in general.
If we decided to PennHIP every breeding Border Collie, and not accept anything over a 0.30 DI for breeding, aside from the other hazards of such a short sighted plan, we might unwittingly be left with a breed of dogs lacking the flexibility needed for their work, or ones with stifle and hock injuries from stress on those joints, since the now perfectly tight hips are no longer helping absorb the stress.
This is what is listed on a PennHIP report regarding breeding recommendations using their results:
PennHIP does not make specific breeding recommendations. Selection of the sire and dam for mating is the decision of the breeder.
NOTE: As a minimum breeding criterion, we propose that breeding stock be selected from the population of animals having hip laxity in the tighter half of the breed (to the left of the median mark on the graph). Higher selection pressure equates to more rapid expected genetic change per generation.
By implementing selection based on passive hip laxity, we expect the breed average DI over the years to move toward tighter hip configuration, meaning lower hip dysplasia susceptibility. The PennHIP database permits scientific adjustment of criteria to reflect these shifts; the average laxity and range of laxity for a particular breed will change over time.
They elaborate on appropriate selection of stock using their results on their website:
Selection Pressure in Breeding
The principal objective of selective breeding is to maximize the pairing of good genes by breeding dogs not affected with (and preferably, not susceptible to) CHD.
For the most rapid genetic change, the breeder can decide to mate only the tightest-hipped dogs within the breed (those with the lowest DI) and then continue to inbreed for tight hips from there. This approach, however, will create increased inbreeding. Founding a breeding program on only a few dogs, and inbreeding on these dogs, would reduce the overall genetic diversity in the gene pool and could contribute to the loss of some desirable traits or lead to the expression of some undesirable traits. This reality affects some breeds more than others. For example less than 5% of Golden Retrievers have hip laxity in the ‘tight-hipped’ range, meaning a DI below 0.30. If one were to require that breeding candidates conform to this standard, 95% of the Golden Retrievers would be excluded from breeding, resulting in a serious reduction in genetic diversity. This breeding strategy would neither be practical nor acceptable to breeders and is not recommended by PennHIP.
So Wall is presenting a straw-man argument–that someone of authority is asking Border Collie breeders to “not accept anything over a 0.30 DI for breeding”–when this is not the case. I haven’t found a single example of PennHIP recommending breeding decisions based on DI, rather they have consistently made recommendations based upon breed-specific distributions. The PennHIP report suggests that anything over median will result in improvement over time and they suggest for more rapid results per generation breeders can consider breeding dogs in the tightest 40% of the breed. Wall’s next paragraph continues the straw-man that someone, mainly PennHIP, is calling for a one-size-fits-all breeding regime:
The problem with using a one factor, “one-size-fits-all” procedure with a disease caused by a combination of factors such as HD, is that all breeds do not have the same percentage contribution of these factors. All breeds of dogs are built somewhat differently, are of different sizes, activity levels, and selected using different standards as the primary criteria. Some of these standards involve performance criteria as a major selection criteria and some don’t. In dogs bred to look a certain way, the selection against such factors as favorable pelvic muscle mass may change the percentage contribution of that particular HD factor in that breed. It’s not surprising then, that adherence to a certain screening and selection technique would be more or less effective in decreasing the incidence of the disease in one breed versus another.
This is sort of rambling nonsense. Dogs are not all the exact same. So what? This observation doesn’t then make dogs so dramatically different that you get to still put Border Collies on a pedestal and ignore endemic problems in the breed or deny that they have pedestrian qualities and faults. Wall is also implying that PennHIP is somehow extreme and thus suspect as a wise tool to use from a population structure stance. I don’t think PennHIP’s suggestions sound radical and inconsiderate of breed health:
To avoid these potential problems accompanying ‘extreme’ selection, PennHIP suggests a more ‘moderate’ approach which goes hand in hand with the PennHIP testing. Particularly in breeds with few or no members having tight (OA-unsusceptible) hips this moderate approach is preferable. In such breeds it is recommended that breeders choose breeding stock from the tightest 40% of the breed (meaning the 60th percentile or better), thereby maintaining an acceptable level of genetic diversity while still applying meaningful selection pressure. By breeding only dogs with hips above the breed average (60th percentile or better) the overall breed average will move toward better (tighter) hips from one generation to the next. Clearly the more selection pressure applied (stricter selection criteria), the more rapid the genetic change.
The PennHIP database ranks each dog relative to other members of the breed making it possible for the breeder to identify dogs whose DI will apply meaningful selection pressure. By applying at least moderate selection pressure, eventually the average of the population will shift with each generation toward tighter hips, increasingly tightening the minimum standard for breeding. By following these time-tested principles of quantitative genetics, ultimately fewer dogs will be at risk for developing OA. Understandably, more rapid genetic change could be achieved by imposing greater selection pressure or by using estimates of breeding value (EBV) from incorporation of the pedigree. These strategies are recommended for the aggressive breeder wishing to achieve the most rapid hip improvement.
So to me, Wall’s argument sounds like she’s attacking the test because she didn’t like the results it showed for her breed. Internalizing success, externalizing failure. Textbook self-serving attributional bias.
Wall’s next argument suffers from mixing data and drawing conclusions that really aren’t supported: she tries to make the claim that Border Collies are more tolerant of laxity than other breeds.
In reality, for our breed, the degree of tightness the researchers at PennHIP have determined in their test breeds will almost certainly ensure no HD, is a value very few Border Collies have. Many, if not most, researchers in HD admit that some breeds are more tolerant of passive laxity than others. In other words, these tolerant breeds tend to have a lower incidence of HD than might be expected from their passive laxity scores. Border Collies appear to be a fairly laxity-tolerant breed.
The first statement is a repeat of the extreme strawman from before; no one is suggestion that we cripple the Border Collie breed by excluding all dogs over a .30 distraction index. Her next statement, that Border Collies are special flowers in terms of their laxity tolerance, is unsupported bunk. Her evidence is mixing OFA stats with PennHIP stats and I just find this exercise suspect given the data she’s trying to use. OFA scores are simply not predicative of much of anything, it’s easy to game that test and PennHIP studies suggest that a full “80% of dogs evaluated as “normal” by the OFA were found to have hip laxity by PennHIP testing that predisposed them to developing hip osteoarthritis in the future.” If we just look at incidence of HD according to OFA, as flawed as it is, Border Collies are not remarkable. They are firmly in the middle of the pack of all breeds and within shepherd breeds of similar structure.
If you want to compare HD rates across breeds you need a scientific study that does not suffer from extreme selection bias the way that OFA statistics do. If you want to make the claim that Border Collies are laxity tolerant you need evidence that compares laxity versus DJD across many breeds and she does not present that for Border Collies.
Although OFA HD statistics are undoubtedly skewed low by pre-submission screening, it’s likely these skewed values are constant across the board for these breeds, making the comparisons between them here valid on a relative scale. The figures in the table indicate the Border Collie is more tolerant as a breed to passive laxity than the German Shepherd.
The first basis of her claim is that Border Collies seem to be more tolerant of laxity than German Shepherds. I’ll let you in on a little secret: all breeds are more tolerant of hip laxity than German Shepherds. The reasons for this should be obvious as the modern fad in German Shepherds is for a crippled roach-back with poor musculature and structure all over their back ends. German Shepherds are outliers as you can clearly see in the above DJD vs. DI. Border Collies are not special flowers by being better than German Shepherds regarding laxity tolerance: all other breeds can claim this advantage. Her table is also not evidence of the claim that BCs are more laxity tolerant than GSDs, only the chart that I’ve included above documents this.
Comparing the Australian Shepherd with the Border Collie, the mean PennHIP DI is very similar, 0.49 vs. 0.50, respectively, yet the reported incidence of HD in the Australian Shepherd is about half that of the Border Collie. Australian Shepherds, as a breed, appear even more tolerant than Border Collies to similar passive laxity scores. It’s also of interest to note the highest mean DI (0.56 of the Shetland Sheepdog) has the lowest incidence of HD, and the lowest mean DI (0.43 of the German shepherd) has the highest incidence of HD of these four breeds. These data suggest that average breed passive laxity scores do not directly correlate with the breed incidence of HD as determined by OFA. It appears the passive laxity score that would almost certainly doom a German Shepherd to HD may not doom a Border Collie with the same DI score, depending on the strength of other factors that may be present in that individual. From these figures, one could also speculate that herding breeds similar to the Border Collie, such as the Shetland Sheepdog and the Australian Shepherd, have further mitigated the effects of lax hips through selection for HD protective factors present in these breeds.
Here’s a masterful attempt at presenting a non-argument as an argument. Non-evidence as evidence. The finding of a non-correlation as significant. “These data suggest that average breed passive laxity scores do not directly correlate with the breed incidence of HD as determined by OFA. ” Well no shit, Sherlock! OFA can’t even claim that their own evaluations correlate with incidence of Hip Dysplasia, so trying to compare what breeds look like in the OFA database versus what their laxity scores look like in PennHIP is a bogus comparison on its face.
Whatever point she’s trying to construct using Shetland Sheepdogs and Australian Shepherds, as breeds similar to the Border Collie, loses weight when you start looking at all the herding breeds, especially ones that are built more like Border Collies. English Shepherds are in the top 30 breeds for incidence of Hip Dysplasia, and I don’t think anyone can claim that Shelties are more similar to Border Collies than English Shepherds are. Cardigan Welsh Corgis and Pembroke Welsh Corgis both have higher rates than Border Collies, as do Pyrenean Shepherds and Old English Sheepdogs. Over all dogs, Australian Kelpies are better than Border Collies, but over dogs born 2006-2010 they are significantly worse. Same with Dutch Shepherds. Bearded Collies, Australian Shepherds, Belgian Malinois, North American Shepherds, Shetland Sheepdogs, Tervurens, Belgian Sheepdogs, and Collies all rank better than Border Collies.
So all the talk of “HD protective factors” is just that, talk. Border Collies, which can claim extensive selective pressure from trial success–arguably more pressure than most or all other shepherd breeds have face, are still nothing special in the middle of shepherd breeds and in the middle of all breeds does not provide any evidence at all for the efficacy of sheep trials to confer protection from hip dysplasia.
So what’s going on with our breed?
The good news is that although it appears Border Collies may, as a breed, have somewhat lax hips, predisposing them to HD, many also have the protective factors such as increased pelvic muscle mass and good hip formation and fit. Increased pelvic muscle mass and good hip formation are factors not taken into account during PennHIP laxity measurements and not always measured accurately in the standard OFA view.
The denial is so thick. She doesn’t like that Border Collies have nothing-special hips according to PennHIP, so she doesn’t like that test. I’m not sure why she likes OFA then, because Border Collies are nothing-special over there too! Regardless of the test, Border Collies aren’t going to be special flowers because they just aren’t special in this regard. You can test-shop all you want, but nothing is going to change the fact that Border Collies have average hips and that makes them predisposed to an undesirable rate of hip dysplasia. There are no more “special mitigating factors” that increase a Border Collie’s protection from Hip Dysplasia vs. the average than there are special complicating factors that increase their incidence of HD. They are average. Not special. If they had more mitigating factors than other dogs, they wouldn’t be average, they’d be exceptional.
And they aren’t. The only thing exceptional here is the degree of self-deception that Border Collie trial apologists are willing to commit to delude themselves that they don’t need to change anything at all about their breeding schemes.
* * *
Comments and disagreements are welcome, but be sure to read the Comment Policy. If this post made you think and you'd like to read more like it, consider a donation to my 4 Border Collies' Treat and Toy Fund. They'll be glad you did. You can subscribe to the feed or enter your e-mail in the field on the right to receive notice of new content. You can also like BorderWars on Facebook for more frequent musings and curiosities.
* * *