The definition of a geological fault and why most dictionaries get it wrong.

The definition of a geological fault, and why most dictionaries get it wrong

One of the most important structures for any mineral explorer to understand are faults.

What, exactly, is a fault? To geologists the answer seems so obvious that few of them (even the writers of many geology textbooks or dictionaries) ever bother with a definition. And when they do, they very often get it wrong. When thinking of a fault, geologists usually have in mind a planar dislocation or fracture where the rocks on either side have slid past each other with the displacement across the fault lying in the plane of the fault itself.  Geological dictionaries and structural geology textbooks (or at least the random selection I have read) often reflect this same misunderstanding and offer a definition specifying a direction of movement along the plane of the fault.  They also frequently describe faults are “brittle fractures”, and add that the observed fault movement must always be “observable”.

The authoritative Glossary of Geology[1] with 36,000 defined terms defines a fault as:

A fracture, or zone of fractures, along which there has been displacement of the sides relative to one another parallel to the fracture(my italics)

The Oxford Dictionary of Geology and Earth Sciences[2] offers this:

Approximately plane surface of fracture caused by brittle failure and along which observable relative displacement has occurred between adjacent rocks. (my italics)

Hobbs, Means & Williams in their well-known (and otherwise impeccable) structural geology textbook[3] provide this definition:

A (fault is a) planar discontinuity between blocks of rock that have been displaced past one another in a direction parallel to the discontinuity. (my italics)

Neville J Price, in his 1964 textbook on rock mechanics (4), describes a fault as:

A fault is a fracture which exhibits obvious signs of differential movement on either side of the plane.” (my italics)

and adds…

 Joints are cracks and fractures in rock along which there has been extremely little or no movement. (my italics)

From that universal source Wikipedia [5]:

A fault is a planar fracture or discontinuity in a volume of rock across which there has been significant displacement (my italics)

Or this one, perhaps the worst of all, which I found on the internet here:

 if rocks on both sides of the (fault) plane have moved relative to each other, parallel to the plane of the fault (faults are shear fractures)…Joints, if there is no component of displacement parallel to the (fault) plane (joints are extension fractures). (my italics)

Finally, my last example, and probably the best (although still flawed) from :

Fault in geology, a planar or gently curved fracture in rocks of the Earth’s crust, where compressional or tensional stresses cause relative displacement of the rocks on either side of the fracture.

Why these definitions are wrong

In the quotes above I have placed the words I find problematic in italics. They are the words that define fault movement as taking place parallel to (or along) the fault plane. They are the words that require the fault mechanism to be brittle (or a “fracture”, which implies the same thing). They are the words that require fault movement to be greater than some vaguely-defined (“obvious”, “significant”, “observable”) and arbitrary minimum value.

If all the criteria of these definitions were strictly applied, they would exclude almost all structures that geologists normally understand by the term fault. They would make it impossible to understand and interpret the multitude of second order structures that occur within a fault, and provide a means of interpreting its history and movement direction. But more importantly, from the point of view of the exploration geologist, these definitions make it impossible to fully understand and predict the emplacement of epigenetic mineral veins.

A fault is a planar zone of rock failure across which relative movement has taken place. The mechanism of that failure may be brittle or ductile. Most faults formed through a combination of both mechanisms.

Any section across a fault, such as an outcrop face, a geological map, or section, is only capable of showing the resolved component of movement on that section. This means that if a pre-fault structure such as a bedding plane is displaced across the fault trace, then, in the general case, the displacement you see is apparent and relates only to marker beds of that orientation. Other beds, with a different orientation, may show different amounts, or even different senses, of apparent displacement. Some displaced beds may show no apparent movement at all on the section on which they are viewed.

The above discussion on apparent displacement applies where the relative movement of the rock masses on either side of the fault have moved laterally past each other along the fault plane. As I have shown, many definitions of geological faults either explicitly or implicitly assume this. But rock masses may also move towards each other or away from each other across the fault plane. This creates a whole new set of geometries.

An accurate definition of a fault must avoid any assumptions about direction of fault movement or the mechanism of fault formation.

Discussion on Fault Movement

Fault Movement Vectors (FMVs) define the relative movement that has taken place between the rock masses on either side of a fault at the end of any given fault movement.  FMVs are the direction of movement of any point on one side of the  fault with respect to any point on the other side. FMVs can be shown as two parallel arrows pointing in the direction of relative movement – one arrow for the rocks on either side of the fault. These arrows may point towards each other. They may point away from each other. They may lie at any angle to the plane of the fault.  Any plane that includes the these arrows is the plane of the FMVs.

If the plane of the FMVs is parallel to the fault plane, then the rock masses on either side of the fault must have moved laterally past each other and the deformation mechanism is that of simple shear If the arrows are normal (i.e. at right angles) to the fault, then the type of displacement across the fault is known as pure shear.  Pure shear may be compressional or extensional. In compression (where the FMVs point towards each other) the rocks on either side of the fault have moved towards each other and there has been a necessary reduction in the volume of the affected rocks. In extension (where the FMVs point away from each other) the rock masses on either side of the fault have moved away from each other, and there has been an increase in the volume of the affected rocks.

Extensional faults are of vital interest to the exploration geologist because the extension provides the space, and the creation of the space the  driving force, for emplacement of epigenetic mineral veins – an important source of ore.

A more detailed discussion on the dynamics of fault formation can be found in my earlier blog post here, entitled: The movement of faults.

Movement of faults fig 1 (2)

The diagram shows a series of two-dimensional slices through rocks affected by different dynamic styles of faulting. The red opposed arrows are the Fault Movement Vectors and indicate the direction of net movement of any point on either side of the fault trace. The sections are all in the plane of the FMVs. Click for a larger, sharper image.

Simple shear and pure shear faults are end members of a continuum of styles of displacement across a fault. Even where simple shear is the predominant deformation mechanism, some parts of the fault will locally exhibit the effects of pure shear. Conversely, in dominantly pure shear structures, there will be zones where the structures observed formed through the mechanism of simple shear. If the FMV arrows lie some angle between 0⁰ and 90⁰ to the fault plane, then fault deformation took place by some combination of simple shear and pure shear mechanisms.

Marker Bed Movement Vectors (hereafter, MMVs) are the paired arrows with which geologists use to decorate their maps and sections in order to indicate relative displacement of pre-fault planar structure across a fault.  These structures are typically sedimentary marker beds, but may be veins or even pre-existing faults. In the general case where the fault is viewed on a random section, and without further data, MMV arrows indicate apparent displacement only. The MMV plane is any plane which contains the arrows and, by definition, is always parallel to the fault plane.  However, the displacement of a single plane across a fault is incapable of fully defining the relative movement of the rock bodies on either side of the fault – only the relative displacement of points on either side of the fault can do that. When a fault is viewed on a random section – such as an outcrop, a mine opening, a geological section or a piece of diamond drill core – marker beds with different orientation may give opposed MMVs across the same fault. The angle which the line of intersection of marker bed and fault makes with the Fault Movement Vectors will determine the amount of apparent displacement on that section. An angle of 90 degrees produces maximum displacement; an angle of 0 degrees will produce no apparent displacement. MMV arrows are only equivalent to FMV arrows in the special case where the section on which the displacement is observed is parallel to the FMV plane (as in the left hand map view of the diagram below).

When viewing a fault on the FMV plane, all marker beds will show the same amount and sense of displacement irrespective of their orientation (see figure below). Conversely, if differently-oriented beds show the same amount and sense of displacement, then the plane on which that displacement is being viewed must be the plane of the FMVs.

Many geologists, including the writers of some geological textbooks and dictionaries, confuse FMVs and MMVs, but FMVs are the more fundamental measure of fault movement.

Plan and Section FMVs and MDVs

A plan view and vertical sections across a strike-slip, simple shear fault affecting differently-oriented marker beds. Red arrows are the Fault Movement Vectors. White arrows are the Marker Bed Movement Vectors. The map view (left) is in the plane of the Fault Movement Vectors: marker beds therefore show the true displacement. Sections AB and BC are planes at right angles to the FMVs: marker beds displacements are apparent only. Vectors. Click for a larger, sharper image. 

Discussion on Joints

The amount of movement that has taken place across a fault can vary through several orders of magnitude – from a fraction of a millimeter to hundreds of kilometers. However, locally developed, brittle fracture surfaces of limited extent across which insignificant displacement has taken place are usually called joints. The suggestion by Neville Price (see his definition above) that there may have been no movement across a joint is of course nonsense: if there had been no movement, there would have been no fracture. By “insignificant” I mean difficult or impossible to see with the naked eye.  However, in practice, a displacement can only be quantified where the fracture affects a marker surface – and sometimes not even then.

Joints are presumably the category of fracture which the lexicographers sought to exclude from their definition of faults by their requirement that displacement be “significant”, “observable” or “obvious”.

Joints form in the same way as faults and should be regarded as a – somewhat vaguely-defined -sub-category of brittle faulting. There is no logic for imposing an artificial division between faults and joints based on some arbitrarily defined amount of movement – what has the resolution of the human eye got to do with rock mechanics?

All faults are caused by stress. In faults with a significant strike extent, the causative force is usually the deviatoric stresses associated with tectonism. Joints can also be caused by tectonic stress but may also be the result of changes of non-deviatoric stresses. Examples are the change in lithostatic stress caused by weathering or rock excavation, and thermal gradients associated with igneous activity. Stress is an abstract force which can only be deduced (if you are lucky). It cannot therefore be used as the basis for a definition of a physical structure such as a fault. That would be putting the cart before the horse.

Joints, as you will have gathered, are notoriously difficult to define. However, even with the  fuzzy definitions that are out there (including mine), the term “joint” remains a useful field term for describing arrays of small-scale brittle fractures.

But joints are micro-faults, nonetheless.

Discussion on Brittle and Ductile Deformation

The method of deformation that enables faults to form, and movement to take place across them, can be either brittle or ductile, or, more typically, some combination of brittle and ductile. Ductile deformation is promoted by high temperature and confining pressure, but if rocks are sufficiently incompetent they can deform in a ductile manner at almost any temperature or pressure. Most faults show evidence for both styles of deformation either at different places within the same fault zone, and/or at different times during its formation. This is particularly true for large-displacement faults which typically affect a range of rocks with different physical and chemical properties, and are a composite of episodic movements that took place over a long period of time.

Where the dominant deformation mechanism is brittle, faults typically consist of tabular arrays of close-spaced, sub-parallel anastomosing fractures separating slices of lesser deformed, or unreformed, rock. This pattern is fractal in that it can occur at all scales from that of a regional map to that of a microscope slide.

Where the dominant fault deformation mechanism is ductile, the strain is more uniformly and smoothly distributed across across the width of the fault.  Ductile faults are often described as ductile shear zones, but they are faults, nonetheless.

So here is my definition of a fault:

A fault is a restricted tabular zone of high strain with relative displacement of the rocks on either side.



[1] J A Jackson & R L Bates (eds), 1980: Glossary of Geology. Published by the American Geophysical Institute, 2nd Edition, 1980.

[2] Michael Allaby 4th Ed. 2013 online version. DOI: 10.1093/acref/9780198839033.001.001

[3] Hobbs B E, Means W D & Williams P F, 1976: An outline of structural geology. John F Wiley and Sons, 571p.

(4) Price, Neville J: 1964. Fault and joint development in brittle and semi-brittle rocks. Pergamon Press, 176p.

[5] Accessed July 2020



Comments are closed.