Skip to Main Content

LRAC Legal Research Class Site (Gotschall)

Introduction to Searching

As mentioned above, there are two ways to search on sophisticated legal research systems such as Westlaw and Lexis – natural language searching and terms and connectors searching.

Terms and Connectors Searching Tutorials

Westlaw Tutorial

Lexis Tutorial

Terms and Connectors Searching

Formulating a terms anconnectors (aka advanced or Boolean) search is basically a two-step process:

  1. Choosing useful keywords (aka words or phrases)
  2. Choosing suitable connectors to string your terms together

The quality of your search, and therefore your results, depends on your skill in both steps!

Click here for a video about choosing useful keywords.

Choosing Useful Keywords

Terms and connectors searching involves using keywords (aka words or phrases) to search for concepts. In this context, a concept is a thing, person, place, cause of action, defense, idea, etc. etc. which can be described using a variety of different words or phrases.

Steps to Identify Useful Keywords

  1. Identify a concept of legal or factual significance from your issue/fact pattern
  2. Think of keywords to express that concept that you can use in a search

Basically, you just ask yourself, what are the different words/phrases that a JUDGE might use when writing an opinion, a LEGISLATOR might use when writing a law, or an AGENCY might use when writing a regulation.


Examples

  • Concept: Very short and young person 
  • Keywords to express the concept: child, kid, minor, girl, boy, infant, toddler, baby, rugrat

  • Concept: Police beating up someone for no legitimate reason
  • Keywords to express the concept: police brutality, excessive force, police misconduct

  • Concept: Whatever happened could not have happened without negligence because "the thing speaks for itself"
  • Keywords to express the concept: res ipsa loquitur (Note: Sometimes for very specific terms of art, there is only one way to say it) ​

Choosing Useful Keywords Continued

There are acronyms such as TRAPP to help you think of concepts/keywords of legal or factual significance from your issue/fact pattern.

Things involved

R - Relief sought

- Causes of Action or Defenses

P - Person or Parties involved

P - Places

Terms and Connectors Search Examples

The following are examples of when it is useful to use terms and connectors searching.

Click here to watch a video about the uses of terms and connectors searching. (Note: Long! Speed it up...)

Trying to Find All Documents on a Legal Issue Example

You are interested in cases involving a police dog biting a person who was just standing around minding her own business. If, after running a natural language search, you are not sure you found all the relevant cases, you can try a terms and connectors search.

Natural language search example:

police dog bit bystander

Terms and connector search example:

police OR officer OR sheriff OR drug OR agent OR "law enforcement" OR unit /45 dog OR canine OR doberman OR shepherd /35 bit! OR attack! OR injur! OR kill! OR maim! /35 bystander OR innocent

Search Within Results Example

You ran the following natural language search: police dog bit bystander fleeing suspect. If you use the Search within results field to search within your natural language results, the search is automatically a terms and connectors search.

Find Every Occurrence of a Terms in a Database Example

You want to make sure that you are familiar with every Arizona law about dogs, so you run the following search to retrieve the word dog (or dogs, dog's, or dogs') anywhere in the text of the Arizona Revised Statutes.

Purpose of Terms and Connectors Searching

Natural language searching works really well these days! So, what is the purpose of terms and connectors searching (and OTHER case/statute finding methods)?

  1. Sometimes saves you the time of having to look through a lot of natural language search results
  2. Helps you locate all documents in a database on a specific topic
  3. Helps you find a needle in a haystack
  4. Helps you make definitive statements about how many statutes/cases/etc. exist on a given topic
  5. The Search Within Results field defaults to terms and connectors searching

 

OR Connector (Westlaw)

Click here to watch a video about the mechanics of terms and connectors searching. (Note: Long! Speed it up...)

The OR connector is used to retrieve documents in a database that contain at least one of the specified terms.

dog OR canine = all documents with dog or canine 

dog OR canine OR shepherd = all documents with dog or canine or shepherd

On Westlaw, a space between terms operates as an OR. You can use the actual word OR too.

officer police = officer or police

officer OR police = officer or police

Phrases (Westlaw)

On Westlaw, a phrase is enclosed in quotation marks. (Note: Remember that a space between words is interpreted as an OR.)

"state trooper" = state trooper

state trooper = state or trooper

You can use phrase searching to search for terms of art, proper names, or adjacent terms you would normally expect to occur as a phrase.

"University of Arizona"

"res ipsa loquitur"

"summary judgment"

Singular and Plural (Westlaw)

On Westlaw, the singular form of a word retrieves the singular, plural, and singular/plural possessive forms of a word. 

dog = dog, dogs, dog's, dogs'   

The plural form of a word retrieves only the plural and the plural possessive forms. 

dogs = dogs, dogs'

Note: Make it a habit to use the singular form of words so you won't exclude any possibly relevant documents.

Truncation (Westlaw)

On Westlaw, use the exclamation point (!) to truncate your terms (usually verbs) so you will retrieve the different forms of a word. 

     bit! = bite, bites, biting, bit, bitten, b*tch, etc.

AND Connector (Westlaw)

The AND connector is used to retrieve documents in a database that contain ALL the specified terms.

On Westlaw, you can use AND or &.

dog AND canine = all documents with dog and canine  

dog AND canine AND "german shepherd" = all documents with dog and canine and german shepherd

officer OR police OR sheriff AND dog = all documents with officer or police or sheriff that also contain dog

Proximity Connectors (Westlaw)

Proximity connectors are used to specify how close together terms must appear in a document to be retrieved by a search. On Westlaw, you can use /S, /P, and /# (any number between 1 and 255).

/#     police /3 dog = documents with police within three words of dog

/#     police /15 dog = documents with police within fifteen words of dog

/S      police /S dog = documents with police within the same sentence as dog

/P      police /P dog = documents with police within the same paragraph as dog

NOT Connector (Westlaw)

The NOT connector is used at the END of a search to retrieve all documents in a database that do not contain specified terms/phrases. 

On Westlaw, you can use BUT NOT or %.

police /S dog = documents with police within a sentence of dog

police /S dog BUT NOT bit! = documents with police within a sentence of dog that also do not contain bit!/bite/bites/biting/bitten

Running an Terms and Connectors Search (Westlaw)

Remember that natural language searching is the default search method for Westlaw. If you type in a string of terms or phrases, or some terms and phrases separated by ANDs or ORs, the system automatically runs a natural language search.

Natural Language Search Examples

  • police dog bite
  • "german sheperd" dog bite AND officer
  • police AND dog AND bite       

To let Westlaw know you want to run a terms and connectors search, you have to do one of the following:

  1. Type adv: in front of your search
  2. Include a proximity connector (/5/S/P) or a truncated term (!) in your search
  3. Click on the Advanced link on the right under the search bar and type in your search
  4. After a natural language search, narrow your results using Search within Results (which interprets any search as a terms and connectors search)

Terms and Connectors Search Examples

  • adv: police AND dog OR canine and bite
  • adv: police /10 dog OR canine /S bit!
  • police /10 dog OR canine /S bite (no adv: required since the search has a proximity connector)
  • police and bit! (no adv: required since the search has a truncated term)

Terms and Connectors Searching - Keep It Simple - Westlaw

Although it is possible to formulate elaborate searches, most research needs are met by relatively simple searches.

Many searches follow the pattern of: keyword /S keyword /S keyword

Just try to keep a few things in mind as you formulate simple searches.

Westlaw Example

After thinking of some likely keywords:

  • Type nouns in singular form
  • Truncate your verbs
  • Put some connectors between your keywords

If your initial search doesn't work, try some alternative terms for your keywords.

 

Westlaw Tipsheet

Below is a tipsheet which summarizes how to run a terms and connectors (aka advanced or Boolean) search on Westlaw.

Click here for a larger version.

 

Generative AI - Good at Legal Analysis!

Generative AI is actually better at legal analysis than it is at finding/citing authority to support the analysis. That is why ChatGPT-4 performed so well on the bar exam!

Keyword Searching Important for Efficiency

Currently, generative AI is not as good at being right and citing authority for assertions in its responses.

Therefore, even in the generative AI world, terms and connectors searchinging is important because of efficiency and certainly.

Recent Sarah Research Example

Question about asylum:

ChatGPT response:

Claude 2 response:

 

 

 

Terms and Connectors Searching

I got a lot of information that was right, but I didn't know what part was right (or even relevant), so I had to search to figure it out.

Search History:

Search results:

 

The future of AI in Legal Practice!

ChatGPT/Claude 2 etc. are not trained specifically for legal research or writing.

However, Lexis and Westlaw AI products are coming out!! For Lexis, possibly before Thanksgiving. These services will be able to cite the underlying authority!

Lexis AI+

What can it be used for:

Three things it can do:

Quality control of answers: