With a week to go to the next LiL update, there is no better time to make good on my promise to analyze the writing changes to Lessons in Love brought about by the rework done a couple of years ago. To begin this comparison I would like to begin with a bit of an superfluous lecture: it is my near-universal experience that writers simply do not improve over time despite the common intuition being that people will improve at their craft over time. It really, really doesm ake sense that the more one writes the better one would get at writing, and therefore that we can anticipate writers improving. In general, though, I have found that not to be the case whatsoever. Most common examples of this are actually, by my judgement, misunderstandings of a completely separate phenomenon: that our evaluation of media improves the more time we spend with it. This is to say, we will have a harsher evaluation of Volume 1 of a book as it introduces characters and brings us into the world than we will on Volume 10 of a book where we are fully integrated into the story and all the gears are turning. I think people will often mistake this momentum for an improvement in writing quality when, in reality, the writing has not changed, only their perception of it. (More shallowly, a lot of people are barely capable of analyzing/parsing prose itself, and tend to associate e.g. improved visuals and music with improved writing somehow, as if a scene looking and sounding better suddenly makes the writing better too).
As for the writing itself divorced of all other factors, well, I have almost universally seen that authors simply do not improve notably over time. A genius will have tend to demonstrate their innate talent in the first work they ever produce, such that even after decades of reading their work one can comfortably enjoy their debut work: Goethe stepped into fiction with The Sorrows of Young Werther and began composing Faust Part One in 1772 (see: the Urfaust); Gene Wolfe produced The Fifth Head of fucking Cerberus in 1972 and proceeded in swift order to release the Book of the New Sun, widely regarded as the greatest work of modern science fiction; I need not mention Tolkien; Dostoevsky began his rich career with the widely lauded Poor Folk; etc. Inversely, someone who produces shitty early work will probably never improve all that much (and I will spare the flame war which may arise by my listing of examples of poor authors).
In every case where I have read a series and felt that the writing quality has improved significantly, I've checked the credits and found that in reality a new writer was hired or something of the like. Every time I've followed an author for a long time, I've found their quality to be fairly stable with the usual peaks and valleys of chance resonance; even following long-term fantasy series, for example, I rarely judge that an author has improved after the 3 millionth word of prose despite the fact you would really expect that much quantity to have made a difference. And yet, it almost never does; a good author may may get a bit better, but a shitty author or mediocre author almost never raise to greatness. Of course, I am but one man and I have certainly not read every single author to ever live, so I can speak only of a trend as I see it.
This now brings us to Lessons in Love: a very rare (in my experience) exception to this. Selebus did in fact improve as an author, and somewhat dramatically; I find this pretty remarkable and believe that this transformation would be well-worth studying for any would-be writer who finds themselves insecure about their prose. We can use Selebus's modifications of his own writing in the past to glean insights into what techniques are more successful than others. For the purpose of this comparison I will be using a copy of 0.23 that I grabbed from I believe the subscribestar, since it was the last version before the rework. For convenience I shall distinguish them as "LiL 0" and "Neo LiL."
However, before I begin, I must make one more superfluous aside: despite being rich with insight itself, I will take the liberty of skipping the opening of LiL 0. It is by far Selebus's greatest embarrassment and I would not like to shine a spotlight on it specifically after he took such great pains to purge it from the game. That's not to say that this entire endeavor of mine is not similarly a slap to the rework's face, but for dramatic effect I will skip the opening regardless; my malice is not so limitless.
Okay, and without further ado, let us finally begin!
An early example of improved writing would be in the first conversation with Ami, when Sensei forgets her name. (For the purposes of this comparison I will take what is stated at face value and not speculate on, e.g., the fact Ami is almost certainly not actually surprised Sensei doesn't remember her name.)
In the original Ami noticing this and introducing her name is a very bland exchange which completes the function of introducing her name swiftly but without much personality. In the rework, it occurs in a much more interesting way: first, "how can you just forget that we've been living together for years" is very natural exposition phrased as a question. Second, "what's next, are you going to tell me youve forgotten my name" plays into the dramatic irony that we the player know Sensei has forgotten her name and in fact her name is ??? as she states this. The pause as Sensei waffles on whether to be honest on this is an early insight into his personality, and then Ami's freakout afterwards is humorous (especially with the visual of her shocked face). To finish matters off with aplomb, Sensei says it's not personal - "probably," prompting another shocked reaction from Ami.
Now, I think the improvement in writing quality is pretty obvious, but since this is a rare subject to be discussed I will belabor the point a bit. The original is wholly unremarkable and almost entirely lacking in humor or character development: it's not actively bad in a way which will earn ire, but it also does nothing good, either, and there is no worse fate for writing than to be completely boring. The rework, however, excels in all areas and manages to tell like 4 jokes in rapid-fire succession. There's a rather popular video that contrasts the old Simpsons with modern Simpsons:
You must be registered to see the links
. The example used is how in Old Simpsons, a joke was never just a one-off; the jokes would have layers, and a funny sequence would build on itself 4-5 twists of expectations, instead of just being a joke and done. We can sort of apply this lens here; in LiL, the original text here... well... wasn't even trying to tell a joke, while the rewrite has humor exuding from every line.
Although I will be only providing a limited number of examples in this post, it's valuable to keep in mind that transformations like the above have taken place all over the narration of the original. Swathes of previously bland, "lazy" even narration is transformed into sequences of multi-layered and original humor. I simply can't cover each one (for example, how superior the introduction of the space war is).
But the improvements were not just in humor. Let us gaze up on example 2.
Again, most people with eyes will be able to distinguish the quality here, but I will belabor the point regardless.
It is here that I must invoke that most dreaded concept in writing criticism: show not tell. This is a concept INCREDIBLY misused by the populace at large and tends to be wielded like a club to criticize a work whenever a reader feels it is being boring in some way. The thing is that 'show' and 'tell' are vague enough words that one can apply them to almost anything: that is to say, someone can accuse anything they want to be 'telling' instead of 'showing' without leaving much room for argument. It is for this reason that 'show not tell' must be invoked with extreme care and a consideration for Chekhov's original purpose in bringing it.
And that purpose was to place greater emphasis on details rather than what the details are describing. Rather than say 'the moon is shining,' describe 'a glint of moonlight on broken glass' so that the reader can infer the moon is shining in a more evocative way. We can say (and this is of course my personal interpretation - there is no singular way of interpreting aesthetic rules in literature) that a reader will tend to feel greater satisfaction and immersion when they are putting many details together to form a whole in their mind rather than simply having the whole thrust into their face, and this Ami quote embodies this extremely well.
In LiL 0, despite how the scene seems to avoid a direct description, we are in reality clubbed over the head that Ami is disturbed by the question about her parents. A wordy, verbose passage descriptions all the various actions she takes, going to such extraordinary lengths as to describe her panicked breaths and wobbly knees. The moment is punctuated by a weak, "Um... Could you maybe ask something else instead?"
In this moment I would suggest there is zero drama or intrigue. We can feel the author telling us to be intrigued, telling us there is drama here... and because it is so obvious, we lose interest; there is nothing really for us to imagine or ponder. Ami herself becomes increasingly a bit of a pathetic figure as she weakly asks him to change his question. (again: not speculating as to whether this is all an act at this moment in time). The idea of alluding to Ami's trauma and the accident with her parents is clear, but executed extremely amateurly.
Neo LiL's superior execution is hardly a question. First, Ami cuts him off before he can even finish speaking, establishing her STRONG AURA. Second, she s simply says 'ask something else," giving no allusion to how she is feeling or what she is thinking behind her mask. When Sensei protests, she again cuts him off and demands he ask something else with no change in her smiling expression.
This is great. THIS is intrigue. We are being 'shown' the detail of her refusing to answer, and from this we start to paint the bigger picture in her mind. The initial characterization of Ami being kind of weak and lame was not strong, but already in Neo LiL one will be starting to sense Ami's latent psychopathy and the damage left by certain events. This dialogue is so good that Sensei (probably deliberately but maybe not) repeated it in a later scene with Wakana and Sensei, where Wakana references 'the jaws of life' and Sensei obstinately says "Choose another metaphor." This is how you get an audience intrigued in characters and wanting to learn more.
Another subtle detail is that he has universally excised bits of narration which imply significant passage of time. For example, the narration says she "focuses for a long time on the distant horizon," but this is completely dissonant with the act of reading the visual novel; we ourselves click through this line in matter of 1 second, and don't see any such deliberation from her, so immersion is hurt. It is a reflection of gained wisdom that Selebus knew to cut these out and in future narration pay greater effort to aligning the reading experience with the reality described in the text.
Now, I will take a brief aside to look at something which is not strictly prose but may be interesting to mention in part.
In LiL 0, after looking at the book of students, the scene ends briskly and cuts to Ami in a bath. In Neo LiL, we find her leaning over Sensei's shoulder ominously. This isn't a matter of prose but in scenes like this we can see a matured ability to simulate the actions of characters and juggle multiple concepts at once; the scene goes from a simple "go to room -> look at journal -> put journal down" to a multifaceted "go to room -> look at journal -> ami was there the whole time -> interaction between two characters" (blah blah). Macro improvements are less interesting in general though (since they're more obvious) so I'll leave it at that.
Anyway, moving on, let us look at the following examples of thoughts (sometimes presented as quiet muttering) being inserted into narration or dialogue:
Here we see a tendency to seamlessly insert the internal narration of heroines within narration. It goes without saying that they are almost all cut from the rework: the thoughts of heroines become largely implied rather than stated. Now, again, I must stress that aesthetic rules are not beyond reproach and to be applied universally; I do not present this with the intention of saying 'and that's why any work which has narration for heroines interwoven into the text is bad.' Rather, in this specific case, we can see a marked improvement from its removal. Why is that?
I'm familiar with a quite remarkable novelist who has written hundreds of unpublished novels purely for his own entertainment. While others may ease themselves into sleep by reading for a few hours, this novelist eases himself into sleep by writing for hours. At one point he described his general process for writing and, while many of his insights were remarkable, the one which has stuck with me the most is his firm rule that he does not know what his characters are thinking at any given moment. He avoids phrases like "He thought about" or "He abruptly remembered" because he wants their internal world to be mysterious and something to think about.
Now, where he goes from here is quite high level, and beyond the scope of this post, but I'll try to convey the significance in a single paragraph. The idea is that thoughts can be seen as 'objective truth,' and the process of obsfucating these truths makes reading interactive; at any given moment, we choose to read certain lines with certain context, we connect contexts to reasons, our mind explains the contexts and reasons, etc. But none of it is definitve: objectively factual. Were we to read the novel again, years later, our minds will be inclined to focus on different contexts and different reasons and produce different explanations. We can forever wonder "What was this character thinking in this moment?" and delight ourselves eternally with different interpretations and contexts. One may say this is where the 'fun' of reading can be, sometimes even for the author re-reading their own work.
Where this applies to LiL is simple: in the modern day of 2025, we can re-read LiL over and over and constantly find new things, focus on new contexts, apply our own relations, etc. And this is possible because Selebus DOES NOT TELL US WHAT CHARACTERS ARE THINKING (as much, anymore). We look at LiL 0 and very frequently the mystique of characters, even Maya, get kind of neutered by these unwelcome looks into the characters mind. The kind of 'objective reality' is forced on us and is much less interesting than the alternative. In Neo LiL, we come across Maya looking at her phone here and ponder what she's thinking; in LiL 0, we're told she's wondering 'if this is all worth it,' and instantly the mystique of the shot is gone.
Again, this is not a hard iron rule. There are points in Neo LiL where the thoughts of characters are explored directly. The important thing is for the author to be conscious of what they are doing. To be aware of the detriments of revealing too much, and therefore to be aware enough to measure the pros and cons when taking the action. The important thing in the case of LiL is not necessarily that these moments were cut; the important thing is the work was improved by them being cut, and Selebus - at some point - gained enough sophistication as an author to notice this.
I might say this is one reason why there is a tendency for writers to not actually improve over the course of their careers. It is not as simple as "oh, my syntax has improved by composing many sentences," or, "oh, my gray matter is now more flexible when it comes to summoning forth words from the dark ether." Improving at writer means grasping and tackling extremely high-level concepts such as the formation of nebulous contexts and reasons in the mind of the reader as they tackle an opaque text. If someone is not familiar or comfortable with these concepts when they first pick up a pen, it is not entirely likely they will suddenly discover them midway through a novel. It is more likely that one who does not appreciate this to begin with will never understand. And it's not like most writers seek out educatory texts on these matters, I think; I'm familiar with authors starting out in workshops and getting feedback from editors, but it seems to me most kind of just skate by on raw intuition once they get their foot in the door, since, after all, that was enough to get their first work out there. (This is all speculation though. There's like a million writers in the world and I have read a fraction of them; why the writers I am familiar with do not tend to improve significantly cannot necessarily be extrapolated across the entire population).
Moving on, the next point is a rather interesting one: the general rising of Sensei's intellect.
Essentially, almost every time where in LiL 0 Sensei had some thought like "I wonder who she's talking about" or "I wonder what that's about" regarding something obvious, Neo LiL just flatly changes so Sensei correctly identifies the subject. This is a highly specific subject, but one interesting to analyze from a variety of angles.
First, though, I would like to decry the temptation to view it through a lens of 'competence porn' as the trope goes. Consistently one of the more laughable lenses with which to view LiL is that of a bildungsroman wannabe where Sensei just needs to follow the hero's journey and end up having grown into a stronger, wiser man; with no offense, the hardest I've laughed is reading one user describe how they could not wait for Sensei to power up like Simon from Gurren Lagann to 'save all the girls' in the end. The comedy of this very concept was depicted in a recent event where Sana hallucinates Sensei coming to rescue her for her delusions, though I would not read too deeply into that myself and use it as my own bludgeon to silence oppositional discourse.
Rather, what I mean to say is that we should not view Sensei suddenly growing more perceptive and intelligence here through the lens of 'sensei becoming more competent' and therefore a superior character. That is a tempting line of thought but not what I believe is most meaningful here. Rather, there are two main ways I want to look at this.
The first requires invoking something only tenuously related: Knox's ten commandments of detective fiction. The ninth rule is rather interesting, and I will quote it here:
The "sidekick" of the detective, the Watson, must not conceal from the reader any thoughts which pass through his mind: his intelligence must be slightly, but very slightly, below that of the average reader.
Knox's decalogue, of course, is not meant itself to be rules universally applied across detective fiction, much less harem slice of life stories. However, from it we can glean an insight Mr. Knox had an entire century ago: that reading is often more pleasurable when the point of view character mirrors the reader in intelligence or close to it. There is a certain instinctive impulse of disbelief when a PoV character is imperceptive or slow to a questionable degree: in the case of the above, almost every single reader will consider it blindingly obvious that Rin is interested in Chika, and so will only roll their eyes when Sensei does not notice this himself.
And are they wrong to? If we consider why these original lines may have been written, it is not likely they came from a place of genuineness. What I would detect is a motivation to artificially fan intrigue and simplify the plot: delay addressing these subjects and try to bait the reader into being more interested by not confirming their suspicions. These goals, of course, fell entirely flat if that was the idea; such imperceptiveness earns only ire and impatience, not intrigue, and so the text is only improved by their removal.
The second angle is simply that stories tend to be more interesting when characters are depicted in a more reasonable and realistic way (which is to say INTERNALLY consistent, not like 'stories about hyper-active manic pixie dream girls are bad because they are unrealistic). Stories tend to develop in more interesting ways when the obvious is observed rather than ignored, and when characters act reasonably rather than arbitrarily stupidly. We can say that Sensei observing Rin's affection for Chika (among other things: this is one example among many others) and changing his course of action in according to this is more interesting than him playing dumb. In this case, the drama of the story is heightened and the intrigue increased because the events feel more 'on our level' and difficult to predict. (It also helps that matters like this are why I can safely recommend LiL even to individuals with IQs four standard deviations from the norm).
Anyway, that's five examples, so that should do for now. We have seen that by the time of the rework into Neo LiL, Selebus's writing had improved in a multitude of ways: his narration is tighter, richer with humor and intrigue, free of various beginner traps regarding how to depict scenes or thoughts, and so on. There is not a single scene not dramatically improved on the rework. And this comparison didn't even bother to touch on "chu~", which is of course the actual lowest moment of LiL and something I will spare Selebus from being reminded of yet again, except by mentioning it now and encouraging everyone to read it on the PC if they haven't to laugh.
If this post was interesting and you werent bothered by having to flick the mousewheel a few times to skip it, SLAM that like button down below and I will consider writing a followup with five more examples. Although I hyperfocused on a few examples, as mentioned almost every scene received a magnificent glowup and could spawn pages of analysis on their own, so I am not lacking for content here.