This is going to be a bit of a rant, but it’s not against you; it’s against Fowler and many others who make similar claims.

I would not label any religion the oldest in the world because I think it promotes the ahistorical idea that religions have timeless essences that are passed down from their supposed origins. Religions are a little bit like the “Ship of Theseus,” in which a ship has its planks replaced bit by bit. When all or nearly all of them have been replaced, is it still the same ship? If this process happened over the course of some 3500 years or so, with wooden planks eventually giving way to modern materials, would it make sense to claim that the ship the oldest on the seas? It’s a philosophical question that different people might answer in different ways, but I would characterize such a claim as an oversimplification. Even more so with Hinduism, which is more like a fleet rather than a single ship.

Some historians will try to establish breaking points at decisive moments for when we can coherently say that something in the distant past is now religion x. For example, historians of Judaism will often distinguish between “ancient Israelite religion” and “Judaism,” and will cite events like the destruction of the Second Temple as the dividing line (though this is contested). Along similar lines, historians often distinguish between “ancient Vedic religion” and “Hinduism.” Do we start Hinduism with the Rig Veda Samhita, the oldest recorded text of the ancient Vedic religion? Or does Hinduism only start when the Upanisads reject the Vedic religions offerings-for-boons economy as shallow and materialistic? And yet the Upanisads are esoteric texts that are mostly of interest for Indian philosophers and modern educated Hindus—their concerns are far removed from the lives of the majority of people, especially before the advent of widespread literacy. So maybe it makes more sense to say Hinduism started later still, with the popular devotional movements that swept through the Indian subcontinent.

Another breaking point might be during the colonial era, when European scholars collaborated with Indian pandits to learn Sanskrit and read the religious literature written in that language. As a predominately Protestant group, those European scholars’ biases led them to assume that all religions were supposed to be based on ancient texts. Based on their readings of those texts, those European scholars decided that India had a religion and they called it Hinduism. But religion and Hinduism were not indigenous categories and did not have any direct translations in indigenous languages. Most of the people who were not being called Hindu might have identified more readily with a caste and a sectarian movement, often in a specifically localized form. This speaks to another issue that complicates all of this, which is that the narrative of a unified Hinduism often elides the countless overlapping, loosely related traditions that are localized to specific communities even contemporaneously. Nevertheless, the British implemented Hindu as an identity category on censuses while Indian civil and political leaders used it as an identity to rally around as the Indian nationalist movement got underway. Does it make sense to refer to Hinduism before this process happened? This is a matter of debate, but at the very least we’d need to recognize that while a wide variety of texts, ideas, practices, institutions, etc. helped to produce a thing called Hinduism in the modern era, they often had long histories during which they should be understood in their own terms. This probably seems like a cop out, but I would suggest to you that the answer to your larger question about why “Hinduism survived” is implicit in a lot of what I’ve said. What Fowler is calling Hinduism is actually a very fluid and decentralized network of loosely related traditions that we as modern people have lumped into a single category. Over the course of history, parts of this network have been displaced by Buddhism or Islam. Other parts were able to adapt, for example by integrating elements of rival religious traditions to maintain their own appeal. An important issue here is that for much of South Asian history, boundaries between what we now think of different religions were very blurry. Being a Buddhist Brahmin or a Muslim Vaishnava was not at all out of the question under some historical circumstances. Finally, even if it were possible to wipe out every node in the network of traditions we now call Hindu, there was no point in history during which it was politically expedient to do so for non-Hindu religious communities that held more power. Religious institutions that did not match the state religion would frequently experience oppression if they were uppity, but if they supported the regime, they might even still get royal patronage. This frequently happened under the Mughals, at the very least.

This was a hot topic 15-20 years ago, but one of the more recent works that speaks to both the construction of “Hindu” as a category and the localized, fluid religious life that existed historically in India would be Peter Gottschalk’s, Religion, Science, and Empire: Classifying Hinduism and Islam in British India.