You can make a contract on the phone; hell, two people in the middle of a meadow can make a verbal contract that's completely valid (provided there's an "exchange of considerations", and the contract is "fair" - depending on jurisdiction).
The problem with verbal contracts is that it's hard to produce evidence about such contracts, if it comes to enforcement.
Email would be OK, if the agreement were assigned to the custody of some mutually-trusted party. It's no good if the parties to the agreement each save their copy of the text/plain agreement on their own corporate mailserver.
The Statute of Frauds requires some societally-significant contracts to be in writing, and makes oral contracts in those categories unenforceable. The law-student mnemonic is MY LEGS — a written contract is required for agreements concerning: Marriage (e.g., prenups); Year (contracts that cannot be performed within one year); Land; Executor (when the executor of a dead person's estate agrees to pay the decedent's debts out of the executor's own pocket); Guarantor (of another's debt); Sales (of goods for $500 or more). [0]
Nit: Strictly speaking a written contract is a "verbal" contract, because in this context, verbal means, in essence, using words, as opposed to, say, gestures or drawings. ("Verbal" derives from the Latin verbum. [1])
Very, very few signed, written contracts get deposited with a neutral. Parties keep signed “file copies” among their records. For big or important deals, one or the other’s lawyer compiles and shares a “closing binder”. Mostly these are PDFs, on file servers and mail servers.
Practical evidence is not technical proof. Methods of technical proof don’t cover all the potential legal challenges.
The goal is establishing terms before judge or jury. Do that and the court can enforce.
Signed contracts have a signature. This has been an acceptable way to ascertain authenticity of the document for a long time. Signatures are surprisingly difficult to forge.
I have developed my signature so that nobody can replicate it fluently. I also use special one off archival ink that nobody else in the world has.
Emails do not have any default mechanism like that, especially if they only rest on servers of the interested parties. Emails are just bytes and can be easily edited.
I’m curious to know your thoughts on the following counterpoints:
> I have developed my signature so that nobody can replicate it fluently.
Robots that hold and move a pen have been created to forge signatures.
> I also use special one off archival ink that nobody else in the world has.
I’m not sure if you’re being serious or not because this is absolutely hilarious, but in the event that you are, does this have any purpose? You could use this excuse to secretly “poison” a contract that you intend on breaking (by using a normal pen to sign it).
> Emails do not have any default mechanism like that, especially if they only rest on servers of the interested parties.
Gmail accounts are free and support DKIM. If both parties conduct business over Gmail, those messages have irrevocable sender authenticity. This inability to refute the contents of an email spelled disaster in the USA for the Democratic political party during their 2016 presidential campaign.
DKIM signatures are required, if you want to exchange mail with Google. Consequently just about all mailservers nowadays append DKIM signatures.
A DKIM signature attests that some subset of the headers (typically including the From: header) are "authentic". The mailserver cannot attest the authenticity of the sender; it cannot attest more than it knows, which is that the logged-in account is allowed to send mail using that sender address. It can't attest who was using the account.
DKIM is a spam-prevention scheme. It doesn't provide non-repudiability.
Signatures typically appear on printed documents. More and more often, they appear on separate signature pages, which don't include any terms or the signature of the other side. Authentication is not document integrity. Printed documents are also easily edited. Print and PDF pages are easily exchanged.
I've seen contracts where parties initialed each page. Don't ask me why. I don't think I've ever seen a contract of more than one page signed on each page.
I own several bottles of Noodler's ink. If I ever sue on a paper contract, I expect I'll be sending a scan or a photocopy into evidence.
There's nothing magical about pen and ink, new or old. We just have a longer history with it. Everything's just evidence.
Everything a Signaturen does is better done by an digital signature.
It's crazy how one party signs a PDF, scans it, and sends it via mail to be signed and scanned by the other party.
Legally no party then owns a signed copy because they just got send a copy of the signature which is void.
Doesn't matter in court: these contracts are enforced. We have done it. In the end only the invoices really matter.
> Signatures typically appear on printed documents. More and more often, they appear on separate signature pages, which don't include any terms or the signature of the other side.
When you sign multi page document you sign every page. This is standard legal practice.
I live in Poland but I am mostly working contracts outside Poland or for foreign companies.
When I receive negotiated contract for signing it is usual to expect the contract has already been at the very least initialed on every page by the other party. And I always initial every page of the contract whether the other party did it or not.
Signing every page is not required for the contract to be enforceable but is standard practice.
Most of the time when there's a court case over these things, the disputed facts do not include which email was sent at what time and what it said. If the other party does choose to dispute those facts, it's true that you would have a hard time proving that you were in the right without some sort of witness like what you're describing; but usually they don't. (It's neither necessary nor sufficient that both parties trust the witness at the time; what matters is whether the court believes the witness if a dispute arises.)
I'm just happy to see something about contracts here that's actually about contracts -- not about the blockchain things called "smart contracts," which are neither smart (okay, that's debatable) but definitely not contracts.
It's sad, because smart contracts were originally about something much simpler and without blockchains. A smart contract was just a contract which partially consists of portable executable code. What made it "smart" is that, due to the included code, it could "know" about certain formal conditions in the contract.
To make this all concrete, a vending machine offers a smart contract to its customers. The vending machine "knows" how much money has been deposited in it, and that lets it be "smart" about when to uphold its end of the bargain and dispense snacks.
Again, you're all using the word "contract" wrong.
More precisely, you are using it in a way that is fundamentally different from how the law uses it, and I believe that doing so is harmful (i.e. this isn't about being a language prescriptivist, it's about confusing people in ways that will likely mess them up.)
In the vending machine hypothetical, the part that people are calling the "smart contract" is the physical workings of the machine itself; it detects the presence of a bill, it takes it, gears turn, and a soda comes out.
There is nothing here about obligations, promises, who is owed what. It's just a machine. And thus should not be called a contract. The contract is the legal obligations that are carried with it.
I strongly recommend that you read "Smart Contracts and Traditional Contract Law, or: The Law of the Vending Machine" by Rohr in 2019, published in a reputable law-review journal [1]. If literally nothing else, it contains the quote, "smart contracts are not necessarily legal contracts," which appears to be the point that you want made.
But also, on p9, the author points out that "it may be silly to think of a vending machine in this way, but when, for example, somebody is sickened by a defective can of Coke and issues arise relating to warranties that apply to the sales of goods, the legal status of those ubiquitous, small-potatoes transactions matters." The article goes on to critique the "just a machine[,] and thus not should be called a contract" view which you espouse.
Yeah and don't you hate how graphs aren't graphs and trees aren't trees? These people talk about bootstrapping when they aren't even referring to footwear! And always talking about bandwidth as if it's a maximum data rate when the term really refers to the electromagnetic spectrum! Makes me so angry how these people treat language as if it's some fluid thing where homonyms develop.
Wow, downvoters, excuse me for having a little fun!
But since you want to be serious, let's bikeshed this: What word would you rather we use, in place of "smart contract," to refer to the small programs that run on the blockchain?
No, smarty pants. I'm telling you that "smart contract" is a bad term for those things because it is unclear and confusing.
A contract is a legally binding statement describing a promise or promises among parties to execute one or more actions. But they are NOT the actions themselves, again, they are a document relating to the actions.
"Smart contracts" are THE ACTIONS THEMSELVES. They're deliverables, they're payments, they're executions of code, but they're NOT CONTRACTS, by any sense of how we've always traditionally used the word.
Interesting perspective, but it sounds like you're mixing up smart contacts with transactions. A smart contact defines what transactions might happen. Transactions are the actual actions. One contract can run for an indefinant amount of time and cause many transactions.
By this definition, a vending machine is a contract.
Which it isn't. You could put a sign on the machine, contact so-and-so if it steals your money, we will definitely repay you. That would be a contract.
Yeah, the price list on a vending machine is an implied contract defining a monetary transaction for food. So if someone wants to call a vending machine a contract, I'm fine with that.
I'm a lawyer and you shouldn't be. That's my point. I've already seen people immediately make this mistake. "Oh, I don't need a lawyer now because the contract is now on the blockchain." Wrong wrong wrong. If you'd like to do a blockchain transaction thing with or without a lawyer, that's fine, the risk is on you. But, the thing called a smart contract does not fundamentally replace what the lawyer does.
Hahaha be nice, but yes. I think I pay a lot of attention to this as a lawyer because of the whole trope of "lawyers always muddy the waters," and the deeply silly but attractive-to-tech-people idea that you can replace lawyering with technology, like with more "precision" or something like that.
Not gonna happen anytime soon; at the end of the day what lawyers do is work on "when humans being human disagree on things where the stakes are big," not "writing things clearly like a computer."
Why wouldn't that be legal? Obviously you should be using S/MIME or GPG if you're worried about fraud but it's not like hand signatures are all that secure.
I wouldn't think about wet ink signatures as a matter of security, per se.
A contract is only valid as long as both parties stick to it. If there's a disagreement, it's up to a court to figure it out. That's where you get into risk mitigation. If you are comfortable with a court making a determination based on an exchange over email, then great. Courts see email as a valid form of proving one's identity. If you want to go further to reduce the risk that a court invalidates the contract, you can do things like having witnesses, wet ink, notaries, etc. Obviously for a contract with low material value, it's not a big deal. Signing a multi-million dollar contract? People start taking whatever measures they can to ensure things are air tight.
A lot of tech people think risk mitigation is only verifying somebody's identity, but a court could decide a contract is invalid because the person signing it had no idea what was inside. So there's a lot more to it.
Actually you should be using S/MIME or GPG if the other party is worried about fraud. If fraud happens (someone claiming to be you signs a contract), it is typically the other party to the contract that loses, because they did insufficient due diligence to ensure that they are entering into an agreement with the correct partner.
Hypothetically, it could be that some law was worded “in writing or verbally, or via telegram” and it didn’t include other forms of communication.
There is in fact a U.S. law that defines “writing” in some parts of the law to encompass electronic communication in some manner as well. So some federal laws that required certain kinds of agreement to be in “writing”, like, IIRC, exclusive copyright licensing, now permit electronic communication.
Because the document at that link can change.
What can work is if there is a hash of the data, and you hope that the data is still there at the link. Better if you have the actual data that was signed. Which means, we're back to actual documents.
What we SHOULD have, however, is a github for lawyers.
The document at the link is a PDF, so the "the document is changeable" problem could be addressed by downloading the PDF and attaching it to the email agreement to serve as the record copy.
> Why not just include the contract in the email then?
One of the advantages of a fixed reference document is that, apart from confirming that you have that document, you don't have to read it to know what it says — and to have a sufficient degree of comfort that the other side isn't trying to sneak something past you.
In contrast, if Party A (or its lawyer) copies and pastes the contents of a reference document into an email, then Party B will have to read the thing, which will take time and, if a lawyer is involved, cost money in legal fees.
An analog that comes to mind is the INCOTERMS rules for shipping goods: If you agree that the goods will be shipped DDP (Delivered Duty Paid), or EXW (Ex Works), or whatever, then you know what you're getting, without having to read the rule again to be sure the other party isn't trying to slip something in on you.
Another analogy is arbitration: If you agree to the American Arbitration Association's rules for commercial arbitration, then all concerned can go to the AAA Website to get a reference copy.
I see your point here, but I'd counter that it's technologically trivial to check that the text matches a reference. If you already have a copy of the contract that you've read and agreed to, and this email claims to have included the same contract in a standard form, then nobody needs to sit down and read the thing.
> it's technologically trivial to check that the text [of an email] matches a reference
Remember who we're talking about here: Lawyers and business people — too many of whom don't even know how to use Microsoft Word "properly." [0]
(You have no idea how many contract drafts I review in Word where the drafter used 'Normal' style for everything and just bold-faced the headings; used blank lines to put spacing between paragraphs; and didn't know how to do automatic paragraph numbering.)
Hard-copy documents can be fraudulently altered after signature too. The question is, how much more burden do you want to impose on the parties to address the infinitesimal fraction of cases where that kind of fraud might be involved.
EDIT: Each party presumably will have email records with the PDF attachment — and if the attachments don't agree, then the parties' lawyers will be busy. But again, how likely is that to happen, and how much time and money do you want to spend to prevent it? At some point you hit diminishing returns (and OCD behavior).
This reminded me of a great talk (in German) by David Kriesel about scanners not scanning correctly but replacing letters and numbers because of pattern matching (if I remember correctly).
It was a really great talk. Not sure if there is an English version somewhere:
Which has german/english dual audio, but defaults to german (so presumably google ruined the english track on upload). If your video player doesn't support picking audio, you can probably fiddle with ffmpeg to strip out the german track.
It's not trivial. Archive.org intentionally makes it an annoying process to either have material removed or excluded from their index/ing. You also must own and control the domain in question where the content is located, which nullifies the premise here (waypointnda.com is owned by someone else).
It used to be trivial to hide existing content in their index from being viewable (for your domains). They entirely changed that a number of years ago.
A paper contract can easily change as well. Ultimately it just comes down to people telling the truth in court, supported by evidence. You'd be surprised how often legal disputes come down to this - but people tend to be honest about these things, because fraud is a tough line to cross, and fraud on the court is quite another.
Preventing a web page from changing requires continuous significant effort. Changing a paper contract requires significant effort, unless you just, you know, lose it, or drop it in the gutter, or something.
Yes, if what the parties are looking at to sign is the printed web page. But we're commenting on an article about signing contracts via email. If they each print the web page separately, they may print out different versions of it. A more effective measure would be to send the contract in the email in some kind of self-contained format such as PDF or plain text.
What are the necessary components of a contract? If I wanted to create a document storage for contract datastructures, each having their own unique content addressable hash, what are the key features that would be needed to be considered a legal contract? Name/Title, creator, recipients, signatures/acknowledgements and terms? Are their specific terms involved. Any specs on this sort of electronic data?
In the US, generally speaking, the following items are necessary for a contract:
1. Multiple parties
2. Consideration: all parties must get something out of it. You may have an agreement where you don't get something out of it, but you cannot have a contract where you get nothing out of it.
3. Offer, acceptance, and mutual assent. One must make the offer, the other(s) must accept, and all have to see eye-to-eye.
4. Legality: you can't make a contract to perform illegal acts. A contract with an assassin is not legally a contract -- one cannot be found in breach of this contract by a court of law no matter what (though bringing the suit may incriminate you for conspiracy to commit murder).
5. Parties legally able to enter contracts. Minors can't enter contracts.
Or just never sign NDAs since they are almost universally egregious and don't benefit you at all to become party to. 99% of the time when you say no, the other party will just shrug and say ok.
I sign, and require signatures, for non-egregious NDAs all the time.
You will quickly find that at outright refusal to sign an NDA will very quickly limit your business options, especially when negotiating between corporate entities.
NDAs are extremely important for businesses to be able to work together on common projects. I work in finance and if I told one of my customers or vendors that we were not going to execute an NDA, the conference call would end very abruptly.
Also consider if the tables were turned and it was you who had the mountains of IP and other liabilities that would be shielded by an NDA.
I've been asked to sign NDAs. "Hey, i got this cool idea, sign my NDA". The couple times I've done that have been... not worth the effort, and never amount to anything. But they're still some mental overhead I have to keep account of for a while.
"We're bringing you in to project X, and we've agreed to payment terms, and project length. Sign this NDA preventing you from sharing our corporate secrets (with some reasonable definition of 'secret') to work on the project". I've generally no problem with those sorts - they are a part of real project/work/life, and generally money is attached.
I've given up dealing with the first sort, and have "lost out on some big opportunities" (their words, not mine) because of the refusal. I've managed to eke out a living nonetheless.
The problem with verbal contracts is that it's hard to produce evidence about such contracts, if it comes to enforcement.
Email would be OK, if the agreement were assigned to the custody of some mutually-trusted party. It's no good if the parties to the agreement each save their copy of the text/plain agreement on their own corporate mailserver.