When a contract says you’ll “indemnify and hold harmless” another party, you’re promising to cover their losses, legal fees, and liability for specific events spelled out in the clause. If one of those events happens, you pay, whether that means reimbursing a judgment, funding a settlement, or absorbing a direct financial hit. Most courts treat the two phrases as meaning the same thing, so the paired wording usually works as a single risk-shifting promise rather than two separate obligations.1Adams v. Atkinson, Supreme Court of Alabama (2025)
Breaking the Phrase Apart
To indemnify someone is to compensate them for specific losses or expenses when a covered event occurs. The payment can cover legal fees, settlement costs, court judgments, or direct financial losses. A software developer, for example, might agree to indemnify a client for losses if the software turns out to infringe someone else’s intellectual property. If a third party sues the client over that infringement, the developer pays for the defense and any resulting judgment.
A hold harmless provision is a promise not to hold the other party responsible for certain damages or injuries. The waiver before a whitewater rafting trip is a familiar example: you agree to hold the outfitter harmless for injuries that result from the inherent risks of the activity, giving up the right to sue over those risks.
One distinction worth keeping in mind cuts across both terms. A third-party claim involves an outsider suing the protected party, like the intellectual property example above. A direct claim is one the protected party raises against the other side for a breach or loss under the contract itself. Courts in several jurisdictions require the contract language to be unmistakably clear before an indemnity clause reaches direct claims between the contracting parties. If the clause doesn’t explicitly address both, it may only apply to third-party lawsuits.
Do “Indemnify” and “Hold Harmless” Actually Mean Different Things?
This is the question most people searching the phrase really want answered, and the honest answer depends on the court. The majority of jurisdictions treat the two as synonymous. The Supreme Court of Alabama reaffirmed that position in 2025 in Adams v. Atkinson, holding that the phrases carry the same meaning even when they appear separately rather than as a pair.1Adams v. Atkinson, Supreme Court of Alabama (2025)
A smaller number of jurisdictions recognize a technical distinction. Under that minority view, “indemnify” creates a right to recover money after a loss, while “hold harmless” provides a broader release that prevents liability from attaching in the first place. The difference would matter if someone sues you and you want to argue you were never liable at all (hold harmless) rather than liable but entitled to reimbursement (indemnify).
Because the law isn’t settled everywhere, contract drafters routinely use both terms together as a belt-and-suspenders approach. If you’re reviewing a contract, don’t get hung up on whether one word appears without the other. Focus on the scope of the clause: what events trigger the obligation, what losses are covered, and whether any carve-outs limit the protection.
How Much Risk You’re Really Taking On
Two contracts can use identical “indemnify and hold harmless” wording and still assign very different levels of exposure. The scope depends on which of three forms the clause follows.
- Broad form: the indemnitor agrees to cover all liability, even if the protected party is entirely at fault. A subcontractor under a broad form clause would pay for an injury caused solely by the general contractor’s negligence. This is the most aggressive form, and many states prohibit it.
- Intermediate form: the indemnitor covers liability unless the protected party is solely at fault. If both parties share some blame, the indemnitor still pays. The protected party only loses coverage when the loss is 100% their own doing.
- Limited form: the indemnitor covers losses only to the extent of their own negligence. If you caused 40% of the problem, you pay 40%. This is the most balanced version and the only type enforceable in many jurisdictions.
The form matters because it determines whether you could end up paying for someone else’s mistakes. A broad form clause in a construction subcontract could leave you responsible for damages you had nothing to do with. Identify which form you’re looking at before you sign.
Defense Costs Are a Separate Promise
An indemnification obligation and a duty to defend are different things, and confusing them creates real problems. Indemnification kicks in at the end of a dispute, once there’s a judgment, settlement, or actual loss to reimburse. The duty to defend requires the other party to fund or provide a legal defense the moment a claim is filed.
The timing gap is significant. If your contract only includes indemnification without a duty to defend, you might have to pay your own legal fees upfront and fight about reimbursement later. A duty to defend means the other party must step in and cover defense costs from the start, which can easily reach six or seven figures in complex litigation.
Some states, including California, have found that an indemnity obligation can imply a duty to defend even when the contract doesn’t say so. Relying on an implied duty is risky. If defense coverage matters to you, insist on explicit “defend” language alongside “indemnify and hold harmless.” Courts generally read the three words as independent obligations: defend, indemnify, and hold harmless.
What to Check Before You Sign
Before agreeing to an indemnify and hold harmless clause, run through these questions:
- Which direction does it run? A unilateral clause means one party indemnifies the other with no reciprocation. A mutual clause means each party indemnifies the other. If the clause runs only against you, consider whether that matches the actual risk allocation. Vendors dealing with large corporations often sign unilateral indemnities because the corporation has the leverage to demand it; joint ventures and partnerships more often use mutual language because both sides bring comparable risks.
- What triggers the obligation? Look for the specific events that activate the clause, such as breach of a representation, negligence, or intellectual property infringement. Vague triggers like “any claim arising from” the contract can sweep in losses you never anticipated.
- Is there a cap? An uncapped indemnity means unlimited exposure. In private acquisition deals, caps commonly fall between 1% and 10% of the deal value. If you’re giving the indemnity, push for a cap tied to a reasonable figure like the contract value or the fees you’re being paid.
- What losses are covered? Pay close attention to whether the clause includes consequential or indirect damages. A clause that excludes them limits recovery to the direct cost of the problem. One that includes them can reach lost profits and downstream financial harm that dwarf the direct loss.
- Are your own mistakes excluded? Most clauses carve out losses caused by the protected party’s gross negligence or willful misconduct. Without that exclusion, you could be paying for problems you didn’t cause.
- Is it backed by insurance? An indemnity promise is only as good as the indemnitor’s ability to pay. Many contracts require specific insurance coverage and name the other party as an additional insured, giving direct access to the policy rather than depending on the indemnitor to write a check.
- How long does it last? Survival periods commonly run from 12 months to several years. Obligations tied to fundamental matters like ownership or authority to enter the deal often survive indefinitely, and tax indemnities often survive until the relevant statute of limitations expires. If a claim is raised in writing before the survival period ends, it typically remains valid; miss the deadline and the right to indemnification can be permanently extinguished.
When the Clause May Not Hold Up
Courts won’t enforce every indemnity clause as written. Several doctrines can void or narrow these provisions, which means the words on the page don’t always decide the outcome.
Approximately 45 states have enacted anti-indemnity statutes, primarily targeting construction contracts. These laws void indemnity provisions that force one party to cover losses caused by the other party’s negligence. The strictest versions prohibit any indemnification for the protected party’s negligence, whether sole or partial. Less restrictive versions only block indemnification for the protected party’s sole negligence. The practical result is that broad form indemnity clauses are unenforceable in most states for construction work, though the specific restrictions vary and you shouldn’t assume the clause is void without checking the law in your jurisdiction.
As a general principle rooted in the Restatement (Second) of Contracts, a contract term that exempts a party from liability for harm caused intentionally or recklessly is unenforceable on public policy grounds. An indemnity clause will not shield you from the consequences of fraud, assault, or other intentional torts, even if the language looks broad enough to cover them.
Courts in most jurisdictions also construe ambiguous indemnity provisions against the party seeking protection. If the clause doesn’t clearly spell out what’s covered, the protected party loses the argument. Many courts require express, unambiguous language specifically referencing the protected party’s own negligence before enforcing a broad form clause.
Reading indemnity language carefully feels tedious until you’re on the wrong end of a seven-figure claim. The few minutes spent understanding what you’re agreeing to is cheap insurance against learning these distinctions the hard way.