# “I Attack” Is Not an Action
*Created:* 2026-10-27, *Modified:* 2026-11-01
![[I Attack Is Not an Action (cover).png]]
*Grab icon © Lorc / game-icons.net (CC BY 3.0)*
This started as a disagreement on X about whether combat in **Bastionland**-style games is boring, and turned into the clearest case for fiction-forward fighting I've read in a while. [김청구](https://x.com/Blue_sphere_a) argued that it only *feels* boring when nobody bothers to think about how they're attacking. I asked for a worked example, and he gave me two.
## Embrace the Abstraction

> "In vanilla **[[Dungeons and Dragons|D&D]]**, the 'attack roll' abstracts the attack action a character takes across that round. To say 'I attack' is for the character to make the best decision 'on their own.'
> "Games like **[Bastionland](https://www.bastionland.com)** and **[Cairn](https://cairnrpg.com)** remove the attack roll, and in doing so hand the choice of *how* you attack back to the player."

> "'Attacks always hit' can only ever be a crude and imprecise description, because in **Bastionland** an attack that doesn't strip all of the target's hit protection has been 'defended against'…
> "If combat in **Bastionland** isn't fun, that's because you haven't thought hard enough about how you're attacking."
I think this would be an excellent time to stick in an example of both what you think of as the failure mode of not thinking it through and what that looks like, and what successfully thinking it through and what that looks like would be. I love concrete examples because they really tell us what you think and how you're thinking about it.
## Four Questions, Superior Outcome

> "Honestly, I think the point where Bastionland-lineage games fail at handling a fight is every single moment you just say 'I attack.'
> "How? With what? Where? To what end? You have to think about the means, the method, the goal, and the intent — all of it.
> "I like to explain this by analogy to the example in **[[Blades in the Dark]]**."

> "Picture two gangsters in a standoff, and the player's character 'Bill' swings at the other boss's jaw. What if you treat that simply as an 'attack'? Then it becomes a moment out of a video game: you attacked, so they attack, turns go back and forth, yadda yadda yadda.
> "There are far more interesting ways to handle the situation."

> "Why swing at the jaw? Because you want them to stagger. Why do you want them to stagger? Because they keep talking over you. So the intent isn't to kill them, it's to take hold of the situation? Right.
> "Which means it isn't just an 'attack' any more. It's a kind of escalated threat. And so you can't treat it as an attempt to deal damage."

> "So did they dodge the punch? Roll damage. If it beats their HP, then you've turned their jaw for them. Oh, so it did. Then the leftover damage comes off Will, and they make a Will save. Oh, they failed. Your punch rattled them — 'this guy is completely insane.'
> "A completely different situation."
## Murderous Intent

> "The same holds in a fight with 'murderous' intent. You can go further than 'I stab them with a knife.'
> "Are you trying to kill them in a single blow? Then roll as usual. See if it beats their HP, subtract from Str, and if they fail their save, they go down.
> "Oh, you're not? Then it gets more interesting."

> "What the other guy is doing is charging at you with a spear, trying to run it through your chest.
> "You want to make him drop the spear? Nothing stopping you. Sliding your blade along the shaft to cut his fingers, that kind of thing?
> "Good. Then roll damage first. Let's see if you can get through his defence."

> "Right, that's damage over his HP. But cutting a finger isn't an attack that 'puts someone down,' so I'll say none of it goes to an ability score. Instead he makes a Dex save. If he fails, he pays the price. Losing the use of one hand — is that enough?
> "Good. Then that's how we'll do it.
> "Same procedure, far more interesting causality."
## Tactical Infinity

> "The fewer rules the game carries, the more freedom the GM and the players get to imagine and realise a sharper image of the game world. Reflecting that intent becomes the facilitator's duty. Tactical infinity is, literally, a procedure for finding the answer in the game world itself rather than in the rules. That's where I find the fun in these games."
## Mox Slips It In From Behind
All right, see, now you've given me an opportunity to give you an example within a game that I really like, **[[Ironsworn - Starforged|Starforged]]**, which has some commonality with **Bastionland** because they both are over in the indie space, but if anything, **Starforged** is a little further out even than **Blades in the Dark**.
Because I love to write examples, we're going to start with the same thing as you originally did, and I'm going to use **Lonelog**[^1] notation for putting the mechanical bits into the text because my usual block embedding just looks a little bit ugly for it.
First, we need to open by saying that **Starforged** is very much a narrativist game, which means that it is fiction forward. The fiction leads the mechanics. I believe that **Bastionland** probably runs better when you approach it from that direction.
What does that mean? It means we have to establish what's going on in the fiction, and the mechanics get triggered by that action rather than the fiction being generated by pulling the levers and pushing the buttons of the mechanics.
So let's start with our fictional framing. We've got two gangsters in a standoff, one of whom is our player character. Our stats define what approaches are most likely to work out narratively in our favor, though they don't necessarily define what the character is good at.
You can be good at things and have the universe really hate you when you do them. In this case, we are going to assume that our character is in fact really good at sneaking and the world likes him when he does it. Pretty good at getting into a close scrap. He's been a gangster for a while.
He's got some bravado to him, which is going to be good for heart. But when it comes to using fast talk or fast thinking to get out of trouble, not so great.
So the numbers land like this:
> [!note] Mox
>
> ```
> [PC: Mox
> | role: gangster — quiet hands, loud mouth
> | stat: Edge 1, Heart 2, Iron 2, Shadow 3, Wits 1
> | asset: Infiltrator, Brawler, Gunslinger
> | gear: pistol, knife
> | meter: Health 5, Spirit 5, Supply 5, Momentum +2
> ]
> ```
> [!example] Infiltrator
>
> ```
> [Asset: Infiltrator|path
> | base: When you make a move to break into a secure site, infiltrate a protected area, or hack or manipulate a secure system
> | base/always -> add +1
> | base/hit -> take +1 momentum
> | base/strong+match -> access is easier than expected; take another +1 momentum
> ]
> ```
> [!example] Brawler
>
> ```
> [Asset: Brawler|path
> | prereq: If you fight unarmed or with a close quarters weapon...
> | base: When you Gain Ground by attempting to disarm, trip, shove, grapple, or stun your foe
> | base/always -> add +1
> | base/hit -> take +1 momentum
> ]
> ```
> [!example] Gunslinger
>
> ```
> [Asset: Gunslinger|path
> | prereq: If you wield a pistol...
> | base: When you Enter the Fray by facing off against your foe (+heart), or by preparing to act without tipping them off (+shadow)
> | base/always -> add +1
> | base/hit -> take +1 momentum
> | base/strong+match -> you may immediately take a shot (without making a move) and mark progress twice
> ]
> ```
I decided to give you the text of the assets involved because they are rather important. The way assets on a **Starforged** character work is that they give you something you wouldn't normally be able to do mechanically, because of a state of affairs in the fiction. For instance, Infiltrator is a path that lets you take a +1 to the applicable roll and if you get a hit, add +1 momentum.
Gunslinger comes into play if you wield a pistol and you're about to Enter the Fray, which is a move triggered by getting involved via the fiction.
(Am I overexplaining things? Probably, but I want everybody to understand what it is they're reading.)
Infiltrator and Gunslinger may be particularly relevant to the current situation. In fact, we probably got there by Mox being in the midst of a break-in against another gang, blowing a roll, getting discovered — here we are. He's face to face with another hard-eyed man-killer.
We don't actually need to create stats for this guy. We just need to know what the situation is. Fictionally, he's just come around a corner and ended up face to face with another guy, probably armed, though we don't know quite yet.
In fact, we can do one of my favorite things and ask the Oracle: is this guy expecting action? I don't have any reason to believe it's one way or another, because we don't have enough of the fiction that led up to this, so I'm just going to say it's 50/50.
> [!note] Ask the Oracle
>
> ```
> ? Is this guy expecting action? — 50/50
> tbl: Fifty-fifty d100=82 -> No
> ```
This guy was not expecting us to come popping around the corner. It was just bad luck on both our parts. The smart thing to do here is to look around and see if we can find a place to duck out of the way.
We could stop here and ask the Oracle if there's any place to duck out of the way, but I don't think that's narratively interesting. Instead, let's just assume there is, because that's the more interesting possibility. Moxie's trying to keep his head down and let this guy go by without catching him. He slips behind a convenient inset doorway, hoping the thug just doesn't look his way.
That triggers the Face Danger move, because we're trying to stealth our way into this place. We're going to be rolling that with +shadow.
I wasn't actually expecting the Infiltrator asset to come into play here, but it definitely does.
> [!done] Face Danger
>
> ```
> @ Face Danger — slip into the inset doorway and let him pass (+shadow)
> d: Action=6+Stat=3+Adds=1=10 vs Challenge=4,7 -> Strong Hit
> (note: Infiltrator — +1 to the roll, +1 momentum on the hit)
> => You are successful. Take +1 momentum.
> [PC: Mox|Momentum+2 -> 4]
> ```
As much as I can't believe it, and you probably can't believe it, Mox successfully tucked himself away behind that little niche, and the thug kept on walking. This puts Mox in a better position to deal with something in the future because his momentum has increased. Good. There's no way that Mox will let this guy keep on heading down the hall and maybe stumble on him again later.
We need to take him down and put him out of the way. Mox slips out behind him and prepares to give him a good grab from behind and a headlock. I'm going to say that the fictional action of slipping out with the intention of coming up behind him unseen triggers the move Secure an Advantage.
We're going to continue using +shadow for this because the sneaking is pretty important. Let's see how that shakes out.
> [!warning] Secure an Advantage
>
> ```
> @ Secure an Advantage — slip out behind him, unseen (+shadow)
> d: Action=5+Stat=3+Adds=1=9 vs Challenge=9,2 -> Weak Hit
> (note: Infiltrator — +1 to the roll, +1 momentum on the hit)
> => You succeed. Choose one:
> * Take +2 momentum
> * Add +1 on your next move (not a progress move)
> [PC: Mox|Momentum+1 -> 5|bonus:+1 next move]
> ```
We are down to get down, but not necessarily to throw down if we can help it. Mox gets in behind him and tries to lock in the headlock, but that triggers Enter the Fray.
Luckily, we are preparing to act against an unaware foe, so we're still working with +shadow.
> [!done] Enter the Fray
>
> ```
> @ Enter the Fray — slip in behind him and lock the headlock (+shadow)
> d: Action=3+Stat=3+Adds=1=7 vs Challenge=2,3 -> Strong Hit
> (note: Gunslinger off — prereq "if you wield a pistol" not met, pistol holstered; +1 forward from Secure an Advantage, now spent)
> => Take +2 momentum. You are in control.
> [PC: Mox|Momentum+2 -> 7|stance:in control]
> ```
Fantastic. Not only have we slipped in behind him and locked in the headlock, but we're in control. Being in control means that we have options in combat that we wouldn't have otherwise.
This means that we also have to envision our objective and give it a rank, which decides how hard it is: effectively, how many mechanical actions it's going to take to resolve the question. In this case, taking down the thug isn't something we want to spend a lot of time on. It's not that important.
So we're going to go with the lowest challenge rank, Troublesome. Taking down the thug is a Troublesome track.
> [!note] Objective
>
> ```
> [Track:Take down the thug|troublesome|0/10]
> ```
We could, if we wanted, use this advantageous position to try and talk this guy into leaving quietly. We could do that. We could chat with him while we've got him in a headlock, trigger a Compel move, and possibly get him to agree to our conditions.
But I'm going to be honest, I don't really want to try and talk him into it when I'm in a good position. I want to choke him out, but I don't want to kill him. That's important.
You'll notice that our objective doesn't say kill the thug. It says take down the thug. Theoretically, we could do it by killing, but that might lead to repercussions we don't want.
People tend to get pissy when you kill their thugs. We don't want that. We just want to take him down, which means we lean into that chokehold, pulling back really hard, putting our forearm right over his carotid artery and cutting off that blood flow.
We want him to go limp, but this is a big guy and it's probably going to be a bit of a challenge. Not a lot of one, but a bit of one, because we're in control. That action triggers a Strike move. We're assaulting a foe at close quarters, which means we're going to be rolling +iron.
You would think that Brawler would come into play here, and if we had more of its upgrades unlocked, it would, but it only applies to Gain Ground now. That's okay. We don't need it. Let's see what happens here.
> [!done] Strike
>
> ```
> @ Strike — lean into the chokehold, forearm across the carotid (+iron)
> d: Action=1+Stat=2=3 vs Challenge=1,2 -> Strong Hit
> => Mark progress twice. You dominate your foe and stay in control.
> [Track:Take down the thug|troublesome|6/10]
> ```
Oh, this guy is going down. We didn't get a great chokehold on him at first, but he had a bad day with those challenge dice of 1 and 2. We didn't even need to use momentum. From 0 boxes on the track to 6 boxes on the track in no time at all. (Strong hits are a hell of a drug.)
We could try to close him out here after a single action. It's better than 50% that we pull it off, but we really need this guy to be out of the way and things not to get more complicated. There's stuff left to do.
We're still in control. We're dominating the guy. We're going to do the classic action movie move of leaning back against the wall and arching our back to take him off his feet and bring his throat down even harder against our arm.
Which triggers another Strike move.
> [!done] Strike
>
> ```
> @ Strike — lean back against the wall and arch to take him off his feet (+iron)
> d: Action=6+Stat=2=8 vs Challenge=1,2 -> Strong Hit
> => Mark progress twice. You dominate your foe and stay in control.
> [Track:Take down the thug|troublesome|10/10]
> ```
This is the hottest my dice have ever rolled in my life, and I can't explain it. Usually when I'm doing an example, I will end up dying from tripping on a rat and landing eye first on a coat hook.
I don't really understand how to respond to actual success. This is crazy.
But since the thug is well in hand and just about completely done, one last squeeze of the carotid should do him just fine. That triggers Take Decisive Action, which will resolve our track one way or another.
> [!warning] Take Decisive Action
>
> ```
> @ Take Decisive Action — one last squeeze of the carotid
> d: Progress=10 vs Challenge=10,2 -> Weak Hit
> => You achieve your objective, but not without cost.
> tbl: Take Decisive Action d100=52 -> Victory is short-lived: A new peril or foe appears
> tbl: Action d100=51 -> Guard
> tbl: Theme d100=48 -> Innocence
> ```
Remember how I was just talking about my luck with dice? Yeah, it had to happen eventually. The thug slides down to his knees against Mox's chest and then flops over limply in the hallway.
He is out of action for a while, and when he comes to, he's going to have a hell of a headache. We prop him up in the niche that we were just hiding in to try and give him some semblance of sleeping on the job instead of being taken out by an infiltrator.
But notice that the result of Take Decisive Action was a weak hit. We have to introduce some complication. The move itself has a table associated with it, and we decided to roll on it rather than just pick something off of it.
Victory is short-lived. A new peril or foe appears. Because I like to take inspiration from the system, not come up with everything off the top of my head, I decided to roll on the action and the theme oracles, which gave me Guard and Innocence. I have a terrible feeling here, but it's a brilliant idea.
Mox takes the headset radio off of the thug's head and puts it over his ear just in time to hear a guard make his hourly check-in: "East Wing Actual to East Wing Three, check in on status. How's it going out there?"
Our man Mox looks around and notices that yes, this is the East Wing. He looks down at the guy he just took out and mutters to himself, "What are the odds that this guy is Three? Pretty much 100%."
"Everything's fine here. Just dandy. I've seen absolutely nothing of note. How's it going with you?"
Aside from borrowing heavily from other media, this is actually fiction that could lead to a variety of moves. Maybe it's a Compel where Mox tries to persuade the guy that he is who he says he is. That seems a little disjointed from the move text, but it's defensible.
Maybe we run this like a second combat, except it's an exchange of subtle verbal sparring. We could actually use that to trigger an Enter the Fray move, followed by Gain Ground as Mox tries to probe for what answer the guy's looking for.
Maybe we read this as a straight-up Face Danger trigger because Mox is attempting something risky or reacting to an imminent threat. That's absolutely defensible here, no matter what we choose. We're going to be guided by the fiction and not end up in a situation in which "I swing my sword" is all you need to say or want to say.
The fiction needs to lead, and when it does, even a relatively limited number of moves that can be triggered become a lot more vivid and textural. I could run the interrogation over the radio as a fight using the combat moves and never swing a fist, because the intention is to achieve the objective and we don't care what the mechanism is. Within the fiction, the fiction looks after itself.
I love fiction-forward gaming. It lets me do fun things — cut off people's fingers in combat, convince them to get down on the floor, trip them with a single smooth flick of my staff — that would normally be very difficult to express purely mechanically. The fiction is the thing that triggers the mechanics rather than the mechanics tightly defining the sort of things that you can say in the fiction.
Mox has had an up-and-down day: an unexpected drop-in on a stranger, a bit of asphyxiation play that came out of it, and a guard on the radio who wants to know that everything's fine. Putting the fiction before the mechanics meant that at no point did we say *"I hit him with my sword,"* and every result had real meaning. That's the best thing we can ever aim for.
## A Fork in the Road
The conversation split above, and the two halves stopped talking about the same thing. One branch stayed with the worked examples: the jaw, the spear, the finger, and from there into the **Starforged** scene above. The other left the examples behind and turned into a discussion about what game terms are for in the first place. That second branch is what follows. Here is the whole shape of it.
```mermaid
flowchart TB
R["김청구 — the attack roll is an abstraction"]
C["김청구 — 'attacks always hit' is a crude description"]
Q["me — show me a worked example"]
F["김청구 — how? with what? where? to what end?"]
A["the jaw, the spear<br/>seven posts of worked example"]
T["김청구 — tactical infinity"]
S["the Starforged scene<br/>one gangster, one chokehold, one guard on the radio"]
B1["me — direction: do the thing, then the move"]
B2["김청구 — the book always follows the game"]
B3["me — fiction leads the book"]
B4["김청구 — game terms break immersion<br/>'+1 to hit' is slacking"]
B5["me — terms of art"]
R --> C --> Q --> F
F --> A --> T --> S
F --> B1 --> B2 --> B3 --> B4 --> B5
classDef other fill:#1d1713,stroke:#3a322a,color:#b9ad9a
classDef mine fill:#2b2318,stroke:#c9a25d,color:#f0e6d2
classDef aside fill:#1a1512,stroke:#c9a25d,stroke-dasharray:4 3,color:#c9a25d
class R,C,F,T other
class Q,B1,B3,B5 mine
class A,S aside
```
## The Book Follows the Play

There should be direction in there at some point, which translates to "If you want to do something, do it." That is, if you want to do something that triggers a move, then your job is to simply do what triggers it, and then the move happens.
Without that, much of the direction in **Blades in the Dark** just doesn't make any sense. You don't see where the initiatory action is. It's with a character who then starts to do something. That do something may trigger a move. If it triggers a move and the obvious answer is undesirable for some reason, then the mechanics come into play, and so on, and so on.

> "That's right. I put it this way: the book always follows the game. What happens in the game world comes first; the book — the rules — is nothing but a servant trailing behind it.
> "Players raised on the mechanics of modern games mistake it for the reverse, believing that the game unfolds on top of the book and that the book exists before the game. A dire mistake."

You'll really appreciate the example reply I wrote elsewhere. :)
I think modern games are the place where the fiction leads the books; traditional design puts the fiction after the mechanics. "I swing my sword at him" is boring and triggers a cascade of resolution that ends up getting interpreted (typically by the GM/KP) within the fiction.
I hate that, frankly.
## Terms of Art

> "1) There are times when it just feels wrong for a player to mention a game term at all. It wrecks your perception of and intuition about the world, and makes it play like a video game.
> "2) In game design, too, I like game terms kept to a minimum; numerical bonuses and penalties are best done away with as far as possible. Something like '+1 to hit' is nothing but slacking."

We're talking about two different things here. In the first point, you're talking about immersion, the idea that the character and the player can sort of merge into a single reactive entity during the course of play, and that you get to think as your character during play.
Which is great. Don't get me wrong, a lot of people love completely immersive play. To a degree, I think everyone does, to lesser or greater. But as a goal state for all the time in playing TRPGs, I think it is unreasonable to ask for or demand.
After all, you are playing a game at a table with other people. There is a reason that the game has rules which have particular terms. They're not there to drag you out of the game. They're there to let you communicate to the other people at the table about your intentions, your observations, and about what's going to happen next.
Game terms exist so that we can be clear and concise in our communication about and around the game. It is impossible to avoid them altogether. Now, you could reasonably ask for them to be minimalized, and I'm totally okay with that, but that has certain trade-offs about expressiveness.
You could decide with the other people at the table to play the game such that all you do is narrate what your character does, first person. You can certainly play that way, and you can agree that you never touch the mechanics. Only the GM/KP/facilitator ends up actually pushing the buttons and pulling the levers of the game that you're playing, and they only communicate with you by telling you what happens as a result.
You can certainly play just about any game that way, but it does mean that you require a GM to play the game and that you require them to do basically all the heavy work in description on results. It also means that you take yourself out of any aspect of creating the game or the background as it goes, for the most part, because usually in order to do that, you have to switch play modes into a more authorial voice.
It's a very restrictive way to play the game, even though it is maximally immersive. It's a great way to burn out a GM so that they never want to run a game again, though, and let me tell you from personal experience, but you can do it.
Now, as to your second point, I'm not sure that's necessarily the truth. "+1 to hit" is not necessarily lazy. What it is necessarily is concise. It doesn't require any other conversation about what it actually means. It has a meaning. It's very clear what its meaning is.
It takes only a few words to convey that meaning. You don't have to ponder over it before you go on to do the next thing. It's just "+1 to hit". Is it appropriate as a mechanical expression all the time? Absolutely not.
One of my favorite systems is **[[Loner]]**, and it has no numerical juggling whatsoever. In fact, none of the skills require you to really deal with numbers. The only thing that actually involves numbers at all is the twist track, and that's just tracking a value between 1 and 3. It has no plus ones to hit.
In fact, the only thing it concerns itself with in terms of mechanics is whether you have an advantage in the current situation, whether you have a disadvantage in the current situation, or you have neither an advantage or disadvantage in the current situation. That's it. There are no plus ones, there's no carryovers, there's nothing about that. Everything else is carried by the fiction.
It is incredibly immersive, but it is not for everyone. However, it is also free to have in pretty much any setting that you like.
So I'm going to toss you a link to **[[Loner - Cthulhu!|Loner: Cthulhu!]]**, which you can check out at your leisure through your favorite machine translation system, because I think it might be educational.
For myself, I don't mind the necessary use of game terms in order to communicate things which are related to the game state. If anything feels like a video game, it's when you stay completely immersed in character and the computer/GM has the job of interpreting everything through the state of the mechanics and making all the decisions about what the world does, and all you do is see the interface in front of you, which can never be perfect.
To me, that's a video game. That is what a video game provides to you. In my mind, a TRPG provides you the ability to access other mechanics which computers don't provide for you, the ability to be part of making the world by stepping into the authorial stance once in a while, and the ability and expectation of understanding the rules such that I can communicate using those rules with the other people at the table.
I do believe that the mode of play described by John Harper in **Blades in the Dark** is one of the best ways to express the ideal state of flow in a TRPG. He says that playing an RPG is a conversation between all the people at the table.
Everyone is talking about what it is that the characters do and how the world reacts, and it's important for everybody in the conversation to be able to understand what's important, how important it is, and how it's going to affect what comes next. In a conversation, you don't strictly use technical terminology. You use it when it's important to talk about something specific that's technical.
In a conversation, you don't just talk about yourself, because that isn't a conversation at that point.
It's probably one of the best treatises on philosophy of gaming I've seen since Ron Edwards put out **[[Sorcerer]]** back in the day, but that's a longer story.
If you're playing a game, you do need to think and react as your character from time to time, but you also need to be able to talk about the game part of the game. To do that, you need specific terms of art. Those are going to happen because they are real things that you need to talk about.
"+1 to hit" may not be lazy. It may just be the simplest way to say what's going to happen. Now, it shouldn't be the only thing you say. It should be couched in the fiction that makes it make sense. Why is there a +1 to hit?
Did your character advance in close such that they're within the opponent's reach and have freer choice of action? Did they spend a turn aiming, drawing a careful bead on the target's head, waiting for the right moment? Did they cripple their enemy's leg, bringing them down to one knee, making them writhe in pain, and thus open for the next major shot?
The problem with "+1 to hit" is not that it is game mechanical speak, but that alone it means nothing within the context of the fiction. Put it in the context of fiction, however, and suddenly it means a lot.
I think that is what's missing from a lot of gameplay from very traditional game players. They don't engage with the fiction. They play the game first and ignore the story. Not the story that's written for them, but the story that they're writing by the course of their actions. That's a bigger deal.
[^1]: **[[Lonelog]]** is Roberto Bisceglie's standard notation for solo session logs and it does the thing a game writeup usually can't: it keeps the mechanics inline without wrapping every roll in a block. `@` is what I do, `?` is what I put to the oracle, `d:` is the roll, `->` is what came back, and `=>` is what it cost. Though I have prettied it up a bit for presentation in the garden.