The idea of purebred dogs is a fairly recent concept and a pretty terrible one. We understand the ramifications of inbreeding in humans, so I don’t know why it seems to be such a cognitive leap for folks to understand why inbred dogs are less healthy and often develop undesirable traits (alongside some inbred desirable ones).
I have had many dogs over the years a number of mutts and a number of pure bred dogs and the pure bred dogs have almost always had both personality and health issues to varying degrees. The mutts tend to have a more even disposition and are longer lived.
The "idea of purebred dogs" is maximization of desirable phenotypic traits, and is identical to the idea of artificial selection throughout agriculture and animal husbandry, which has underpinned the sustenance of our civilization for many thousands of years.
The idea that artificial selection is a terrible idea, on the other hand, is a fairly recent one, and it is entirely informed by its political dimension of eugenics as a peculiar form of totalitarian intervention into human life. Inasmuch as it's terrible to preserve purity of dog breeds, this is entirely explained by a problematic breed standard, usually because it emphasizes some aesthetic feature over generic vitality and ability.
There's another terrible (and wrong) idea here, namely that the best way to maximize those generically good traits is through random interbreeding and preferring crosses with high phenotypic and genotypic divergence; that "mutts are the smartest and healthiest dogs" and so on. This, too, is overwhelmingly informed by a misguided generalization of opposition to a certain abhorrent political project. But reversed stupidity is not intelligence, and misguided anti-nazism is not a good rule of thumb for animal husbandry. In reality, it's not a surprise that e.g. the best, kindest and smartest assistance dogs come from breeds we know as reliably helpful.[1]
It's also not a surprise that Labrador Retriever dogs have unreasonable appetite and high obesity rate: we know the exact specific gene that's broken which explains it. I hope we fix it one day, that's the least they deserve. We easily can fix it, too. And this is the state of health issues with most non-decorative breeds.
> We understand the ramifications of inbreeding in humans
When we understand something, we usually can quantify its effects. As it happens, we do understand inbreeding and why it's bad, both in humans and in the general case. It's mostly explained by increased occurrence of homozygous-recessive phenotypes. Basically, most deviations from evolutionarily optimized alleles are suboptimal, recessive, and in the limit, get weeded out by selective pressure (purifying selection). But a very limited gene pool can become saturated with recessive alleles, leading to many specimens who are homozygous-recessive and get the suboptimal trait; and in bad cases, a recessive allele reaches fixation.
Accordingly we have different relevant metrics and notions, e.g. of effective population size [2] (number of effectively distinct breeding units in the mathematical model of this population; more relevant than census size when you have technically large but highly inbred populations – corn, dairy cattle, pugs, Austrian monarchs…) and of minimal viable population [3] (number of specimens sufficient to ward off the drive to homozygosity and eventual inbreeding depression). The typically given safe number is about 500.
Even very small natural human populations (on the scale of tens of thousands of people, a minor ethnicity or tribe) do not have any inbreeding problems, nor get any heterosis aka "hybrid vigor" i.e. benefit from interbreeding and canceling the accumulated inbreeding depression. Observations to the contrary come from extremely bad cases, like people from isolated premodern villages with some 5-10 families moving to the city for the first time in centuries/ever, and their children being markedly healthier in all sorts of ways (that said, even in such cases, improvements in nutrition, education etc. can explain an obscene share of the effect). But these children do not get more (or less) fit grandchildren after they marry even more genetically distant people from another country – they're already free from the curse of inbreeding depression their ancestral populations carried. What drives these effects is heterozygosity, not total diversity.
This is true for reasonably maintained (i.e. without severe recent bottlenecks) non-decorative dog breeds as well. They don't produce super-great mutts. They're fine as they are.
"observations to the contrary comes from extremely bad cases" do you have any reading on that topic?
I don't have any position. BUt the idea of purebred is a recent one. I think the difference is that it's selecting for even minute appearance instead of a specific desireable trait. Which makes the whole breed very similar to one another as to be clones of each other.
Yes, that dog who wins that award is from a dog breed, but that's because the spirit of the world is heavily biased right now towards dog breeds. No one wants mutt let alone train them.
There is enough research on it, but it often looks at the problem through the lens of incest rather than whole community health and trajectory. E.g. for the UK Biobank sample [1]:
> We thus identified 125 unrelated participants (65 males and 60 females) whose genomes are consistent with their parents being first- or second-degree relatives. That represents a prevalence of [extreme inbreeding] ~0.03%, i.e., ~1/3652
> It is worth noting that complex inbreeding loops between second degree-relatives may also lead to extreme values of FROH. However, mating between first-degree relatives remains a more parsimonious explanation of the empirical observations, in particular in a population of European ancestry where such complex inbreeding loops are uncommon
> We quantified the consequences of EI on multiple traits measured in the UKB. We first analysed ten control traits with prior evidence of inbreeding depression4,8,10,13. Those ten traits are height, hip-to-waist ratio (HWR), handgrip strength (HGS; average of left and right hand), lung function measured as the peak expiratory flow (PEF), visual acuity (VA), auditory acuity (AA), number of years of education (EA), fluid intelligence score (FIS), cognitive function measured as the mean time to correctly identify matches (MTCIM) and fertility measured as the number of children (NCh). ...
> As expected, we found that EI cases had a reduced mean in these ten traits as compared to EI controls. More specifically, we found phenotypic means in EI cases to be between 0.3 and 0.7 standard deviation below the population mean (Table 3). Note, that under normality assumptions, between ~25 and ~40% of the population has a phenotype below 0.7 and 0.3 standard deviations below the mean, respectively.
> We also specifically estimated the inbreeding load (often denoted B), which represents the number of loci with deleterious alleles that would cause one death on average if made homozygous3 ... and found an estimate of B ~1.46
And for India [2]:
> A cohort of 408 children (6 to 15 years of age) was selected from inbred and non-inbred families of five Muslim populations of Jammu region.
> We found significant decline in child cognitive abilities due to inbreeding and high frequency of mental retardation among offspring from inbred families. The mean differences (95% C.I.) were reported for the VIQ, being −22.00 (−24.82, −19.17), PIQ −26.92 (−29.96, −23.87) and FSIQ −24.47 (−27.35, −21.59) for inbred as compared to non-inbred children (p>0.001). The higher risk of being mentally retarded was found to be more obvious among inbred categories corresponding to the degree of inbreeding and the same accounts least for non-inbred children (p<0.0001).
The above reasoning explains the near-universal human taboo against this guaranteed way to increase occurrence of homozygosity and run into inbreeding depression – instantly or in a small number of generations. But small isolated traditional communities risk turning effectively incestuous, and some traditions even normalize cousin marriages. E.g. in Pakistan it's a big problem and is researched intensely [3]:
> Tribal and caste systems are deeply rooted in remote areas of Pakistan. Cheema said that the caste system, particularly among the Arain people living in Punjab province, is especially rigid and leads to many inter-family marriages. She said several genetic disorders are commonly found in this community.
Effective population sizes in those endogamous clan societies are orders of magnitude lower than census sizes, they can be in the low dozens for a given group.
If we return to pedigree dogs (in UK again), the overall situation is admittedly not great but there's cause for optimism. Ironically, one of the most stereotypically "fancy" and purebred dog breeds is in the best position: [4]
> For all breeds, the trend was for the rate of inbreeding to be highest in the 1980s and 1990s, representing a major contraction in genetic diversity. Since 2000 however, the general trend has been for the rate of inbreeding to decline to sustainable levels, with some modest restoration of genetic diversity in some cases.
> Rate of inbreeding (ΔF) and effective population size (Ne) Of all breeds with an average of >50 registrations per annum (over each of the seven 5-year blocks, n = 121), five had a negative whole period ΔF implying an apparent overall increase in genetic diversity, and consequently no determinable Ne (Bernese Mountain Dog, Briard, Standard Poodle, Rhodesian Ridgeback and Tibetan Terrier). Of the 116 remaining breeds, the Ne calculated over the period 1980–2014 ranged from 23.8 (Manchester Terrier) to 918.8 (Borzoi). Of these 116 breeds, 68 had Ne of <100, with 29 having Ne of <50.
In wildlife conservation, Ne≥50 is the minimum for short-term population viability (animals are also smart enough to generally avoid mating with close relatives), but isn't considered good by any means. Still, if we adopt some utterly trivial mitigations like not using the same sire to produce 20% of the next generation of pedigreed dogs, we'll quickly eliminate the bulk of the problem.
>The idea of purebred dogs is a fairly recent concept and a pretty terrible one.
It's just that the inbred traits are so desirable. I've had many mutts, and loved them very much. But when you want a dog, and you want to know exactly what you're getting, it's hard to beat a pure bred. Yes they will have problems, but it's a known quantity. And what you get in return is a dog who needs minimal training to behave exactly as you wish.
If you need a dog to perform any kind of specific function, you're 95% of the time vastly better off with a purpose-bred dog than some random mixture.
Plus, a huge fraction of the dogs up for adoption in the US have strong admixture with breeds with specifically undesirable behavior (e.g. biting children or other dogs), which is often causally related to why they're up for adoption in the first place. Adverse selection, yo.
To be fair, this entirely depends on the breed in question. There are some breeds that are known to have weak bloodlines, and some breeds that are better.
It is also dependent on the breeder. Breeders that "cheap" out will be more likely to have inbred dogs since it can cost money for a good stud.
I have had many dogs over the years a number of mutts and a number of pure bred dogs and the pure bred dogs have almost always had both personality and health issues to varying degrees. The mutts tend to have a more even disposition and are longer lived.