I tend to want to lump things together, especially when at times, species that have been made separate can easily hybridize and form fertile offspring.
I think Iām a lumper but itās hard to tell from the inside 
Iām a shameless splitter. āSpeciesā is a human construct anyway, the lines are always going to be blurry, so in my opinion if itās a taxonomic entity we feel the need to talk about, may as well give it a name to make it easier to discuss.
Usually a lumper 
But after all, I think the decision on splitting or lumping should be made so that the resulting taxonomy is more useful for the uses that will be made of it. As paul_dennehy noticed, taxa are human constructs. If two populations or evolutionary branches have a noticeable difference and for most uses it is worthwile to have information about that difference in the taxon name, probably itās a good thing to put them in two different taxa. Although I think the benefit should always be weighted against the burden of associating new names with records using previous taxonomy (and I think this burden is too often overlooked).
The think I really hate is when splitting produces loss of information: for example, imagine weāre going to split Turdus merula in two species distinguishable only from genetics. Until now, almost everyone can tell Turdus merula from other Turdus species. After the split, we risk having millions of Turdus sp. observations, just because no one is able to tell the two species, but we lost the information that those millions Turdus sp. observation are NOT Turdus pilaris, Turdus viscivorus, etcā¦ Yes, we can create a Turdus merula species group, but are we sure that every observer, application, etc.. will consistently use species groups?
There will be a lot less incorrect identifications if species are lumped, instead of delineations being microscopic details. Just saying. 
This is one of my favorite models for understanding human behavior in countless contexts, not just natural history. When you start looking at argument thru this lens, lots of arguments turn out to be based on disagreements over the applicable level of specificity.
It is the case that some people tend very much towards one of these over the others.
My personal philosophy is to try use the degree of abstraction/specificity that seems to be most useful for the specific circumstances.
I am a special case, I would say: higher taxa I lump (domain>ā¦>family; āmacrotaxaā) lower taxa I generally split (genus>ā¦>subspecies āmicrotaxaā). This stems from the fact that I am a creationist and prefer to use a custom baraminological (look it up) system, where macrotaxa do not show evolutionary relationships that I donāt believe exist (case in point, my mammal order Bovimorphia, which includes elephants, horses, cows, hippos, etc.) and microtaxa show speciation and adaptation from two original ancestors (the ancestors and all descendants form a ācreated kindā or ābaraminā). It is better to split within a baramin and lump above a baramin.
One reason Iāve heard from people who split is that it allows legal (or other) protection for sub-populations of a formerly single species. This is because instead of having a single species that is more cosmopolitan, you get an entire population that might be localized to some endangered environment. Thus, youāll be able to claim protection for that population, whereas if it was more cosmopolitan you would not be able to claim it.
Lumper when IDing and splitter when I want an ID?
Lumper when IDing and splitter when I want an ID?
The optimal condition 
Great topic, Iām a Lumper for species and a splitter for varieties within a species. I think most species complex should count as 1 species with lots of unique subspecies/varieties. Of course everything is case by case.
Brassica oleracea, Brassica rapa, Cucurbita pepo, Cucumis melo, Vigna unguiculata & Vicia lens are all examples of single species taxons that contain so much diversity in varities/subspecies. It also just happens that all of these have been domesticated.
Rubus argutus complex, Solanum nigrum complex (both american subcomplex & european subcomplex), Amelanchier arborea/canadensis (North American Eastern) complex, Vicia sativa complex, Malus baccata/asiatica complex and many more are examples that should be merged under to simplify this taxonomic mess. Why must we cause ourselves more headaces (donāt we have enough species as it is - especially when you can hybridize so many of them easily & make ānew speciesā?
There are few cases where species are merged under 1 genus but should be seperated, for example Section Eriomalus of the Malus genus should be seperated into itās own genus because itās phylogenetically distant. Since in the Maleae tribe, inter-generic hybrids are common (The entire tribe could be a single genus) but distinct genus concepts are still useful & many intergeneric hybrids are already documented.
Indeed! 
Species are absolutely a human made concept/tool but whatās the point of this tool if itās not useful right? Especially if I can just hybridize everything (Especially wide inter-generic hybrids), you gonna have wayyy too many hybrid species to split up 
Ah that explains why we have so many species , legality is weird. Altho why canāt you have an endangered subspecies/variety?
I do also wonder how many extinct heirloom/landrace garden vegetables/flowers are there? They also deserve to be protected especially if they were beloved & appreciated for historical preservation (+ might have useful genetics that were lost with time which could help modern varieties). Same reason why wild tomato subspecies/varieties need protection (or at the very least someone should save their seeds to grow them out & share with as many gardeners/botanists all over the world - this way extinction can be avoided for valuable plants).
Very much an uphill battle (source):
Iām splitting the difference and saying Iām a lumptter. Iād love to tell you which side I fall on, but this all went right over my head and I have no idea what any of this stuff means. 
Remember, Iām a graduate of a US public school and learned all I know about entomology by reading Dr. Seuss Discovers Bugs. Iāll go sit in the corner with my jacks while the grownups talk about enteromulogee. 
.
I believe I heard this from a guest in the podcast In Defense of Plants who is involved in the protection of endangered species.
Obviously, thatās not the actual reason for splitting, but he did note that legal protections are for endangered āspeciesā, and thus if one can show that a disjunct population in some habitat that is planned for modification (e.g. building new mall, etc) is an entirely new species (and not simply a subpopulation of an otherwise successful and more cosmopolitan species), then itās more likely that the habitat will be protected.
you can; Australia has plenty of these, where the entire species may not be threatened, but one of its subspecies is formally listed as endangered