You make many salient points. Great post. Some thoughts:
"Throughout the entire history of both Steem and Hive, people have used downvoting as a way to remove rewards from posts and thus to limit the reach they have on the network."
The issue isn't the voting mechanism but this systemic conditioning that posts with high payouts are "good" and posts with low payouts are "bad". This is a direct relic, in my personal opinion, from the launch of Steemit where Ned/Dan upvoted to the tune of many thousands those posts they figured were the ideal (or the most beneficial to themselves due to pre-existing business relations). Posts that don't get high votes or have a mix of up and downvotes are considered bad posts by those who have been here for a long period of time. It is seen more advantageous to comment on a high-payout post than a low-payout.
This has nothing to do with abuse or anti-abuse efforts to curb fraud and exploitation, which represent a very small amount of votes. It's an attitude which was artificially cultivated.
"The original Steem whitepaper explains that negative behaviour by stakeholders with regards voting is disparaged by the potential cost to them financially if their actions ruin the entire network."
I think you're paraphrasing the crab bucket theory but correct me if you're referencing a different part of that whitepaper. If you read the Hive whitepaper, this type of thing was removed. The old Steem whitepaper refers to the use of Steem only as Steemit the blogging site while with Hive, it is far greater than just single-site blogging. There are a lot of points in the Steem whitepaper that deal with the perceived behavior of people all over the world, much of it discriminatory. I would not apply them to Hive.
This theory also saw an unfortunate application of their one frontend with hiding downvoted posts and comments. That was unfortunate (but likely a business decision as you alluded to in another part of your post) and sparked the whole view of downvotes = censorship. Downvotes are not meant to regulate what we see on frontends. They are meant to allow us to express ourselves.
"A free speech network that is perceived to be downvote heavy and that can easily squeeze out smaller accounts who rely on their blogging income to free up their time sufficiently to enable them to produce quality content - will flush itself down the toilet."
You've read too much of Dan's Steem whitepaper. What you're saying here, subconsciously, is that poor users who want to share their opinions are restricted by time because they must make money. If they cannot make money because their opinions are unpopular, then they can't post. You are assuming that the person must work hard during all hours of their day and that they can't leverage their capabilities (their brain) to find other money generating means on Hive. This is the type of assumption that Dan did and that's why I personally found his whitepaper reprehensible and written from a privileged and ignorant standpoint. People's material desires are not greater than their convictions.
I do understand what you likely mean to say, which is poor users may get discouraged when their work is evaluated negatively, particularly if this is their original content and genuine thoughts. But we cannot start bringing ourselves to believe that their convictions are at the hands of the wealthy. A person who believes the world is flat will not stop believing that simply because of downvotes and will keep writing about it. They are here to share their thoughts even if unpopular and they have the capability to educate themselves about Hive and find other means to generate their income through it if they are indeed set on Hive being their primary income source.
Much of what I said is about attitudes, conditioning and the whitepaper in particular. That's simply because I had to study the Steem whitepaper in absurd detail to initiate the formulation of the Hive whitepaper (and it's a topic I can talk about for hours).
"... targeted downvoting based on ideology casts a very negative light on Hive in the wider world"
Let's flip this argument now and forget about those whitepapers. People do use downvotes to express themselves, just like they use upvotes. If they cannot downvote, meaning if they cannot disagree with rewards and acceptance of a concept (with acceptance presumably shown as upvotes) then they themselves are silenced. This is a part of cancel culture which dictates that no one can say anything contrary to what is being said. And more importantly -- another relic -- that having a negative opinion of something should not be conveyed (everything should be positive and if you criticize anything you are out). What I mean to say is that downvotes are part of free speech. We can't eliminate one type of expression because its unfavorable to some. The mechanism behind the blockchain is intended to have both types of votes to allow all users to express themselves through quantitative transactions.
In summary, the issue at hand is attitudes. The break from years of pre-Hive conditioning is a slow process.
RE: When Is a Censor Not A Censor? The Wider Implication of Downvote/Cancel Culture on Hive.